December 2019 Vol. 14 – N. 2 December ISSN 1827-9635 2019 Vol. 14 – N. 2 Acta Herpetologica ACTA HERPETOLOGICA Iscritto al Tribunale di Firenze con il n° 5450 del 03/11/2005 di Firenze Tribunale Iscritto al 70% DCB Firenze Abbonamento Postale - Spedizione in Italiane S.p.A. Poste
FIRENZE UNIVERSITYPRESS Acta Herpetologica
Acta Herpetologica è la rivista ufficiale della Societas Herpetologica Italica (S.H.I.), un’associazione scientifi- ca che promuove la ricerca erpetologica di base e applicata, la divulgazione delle conoscenze e la protezio- ne degli Anfibi e Rettili e dei loro habitat.
Acta Herpetologica is the official journal of the Societas Herpetologica Italica (S.H.I.), a scientific associa- tion that promotes basic, applied, and conservation researches on Amphibians and Reptiles.
Direttore responsabile (Editor): Marco Mangiacotti, DSTA, Università di Pavia, Via Taramelli 24, 27100 Pavia, Italia
Redattori (Associate Editors): Aaron Matthew Bauer, Villanova University, United States Adriana Bellati, Department of Earth and Environmental Sciences University of Pavia, Italy Daniele Pellitteri-Rosa, Università degli Studi di Pavia, Italy Dario Ottonello, Centro Studi Bionaturalistici, Italy Ernesto Filippi, Sogesid-Ministero dell’Ambiente, Roma, Italy Emilio Sperone, Università della Calabria, Italy Fabio Maria Guarino, Università degli Studi di Napoli “Federico II”, Italy Marcello Mezzasalma, Università degli Studi di Napoli Federico II, Italy Marco Sannolo, CIBIO-InBIO, Universidade do Porto, Vairão, Portugal Raoul Manenti, Dipartimento di Bioscienze, Università degli Studi di Milano, Milano Rocco Tiberti, Università di Pavia, Pavia, Italy Simon Baeckens, University of Antwerp, Belgium Stefano Scali, Museo Civico di Storia Naturale di Milano, Italy Uwe Fritz, Museum of Zoology, Senckenberg Dresden, Germany
Consiglio direttivo S.H.I. (S.H.I. Council): Presidente (President): Roberto Sindaco Vice Presidente (Vice-President): Sandro Tripepi Segretario (Secretary): Dalila Giacobbe Tesoriere (Treasurer): Gulia Tessa Consiglieri (Council members): Gentile Francesco Ficetola, Lucio Bonato, Luciano di Tizio
Sito ufficiale S.H.I. (Official S.H.I. website): http://www-3.unipv.it/webshi
Modalità di associazione Le nuove domande di associazione sono esaminate periodicamente dal Consiglio Direttivo; solo succes- sivamente i nuovi soci riceveranno la comunicazione di accettazione con le modalità per regolarizzare l’iscrizione (ulteriori informazioni sul sito: http://www.unipv.it/webshi). La quota annuale di iscrizione alla S.H.I. è di € 35,00. I soci sono invitati a versare la quota di iscrizione sul conto corrente postale n. 62198205 intestato a: SHI Societas Herpetologica Italica. In alternativa è possibile effettuare un bonifico bancario sul Conto Corrente Postale: n. conto 62198205 intestatario: SHI Societas Herpetologica Italica IBAN: IT-54-K-07601-03200-000062198205.
Membership The S.H.I. Council will examine periodically new applications to S.H.I.: if accepted, new Members will receive confirmation and payment information (for more information contact the official website: http:// www.unipv.it/webshi). Annual membership fee is € 35.00 (Euro). Payments are made on the postal account of SHI Societas Herpetologica Italica no. 62198205, or by bank transfer on postal account no. 62198205 IBAN: IT-54-K-07601-03200-000062198205 to SHI Societas Herpetologica Italica.
Versione on-line: http://www.fupress.com/ah Acta Herpetologica
Vol. 14, n. 2 - December 2019
Firenze University Press Referee list. In alphabetical order the scientists that have accepted to act as editorial board members of Acta Her- petologica vol. 14 (2019).
Elenco dei revisori. In ordine alfabetico gli studiosi che hanno fatto parte del comitato editoriale di Acta Herpe- tologica vol. 14 (2019).
Alejandra Fabres; Anamarija Zagar; Antonieta Labra; Ariane Standing; Carlos Taboada; Dan Cogalniceanu; Daniele Delle Monache; Dirk Bauwens; Elena Grasselli; Fabio M. Guarino; Fabrizio Oneto; Federico Marrone; Feist Sheena; Francois Druelle; Giulia Tessa; Hendrik Mueller; Joanne Paul-Murphy; John Sikes IV; Josabel Belliure; Jus- tin P. Lawrence; Kerim Cicek; Marco A.L. Zuffi; Marco Sannolo; Marcos Peso; Mariana Cabagna-Zenklusen; Mattia Falaschi; Michel-Jean Delaugerre; Miguel Carretero; Naeimeh Eskandarzadeh; Philippe Geniez; Raoul Manenti; Ricar- do Montero; Roberto Sacchi; Rui Rebelo; Santosh M. Mogali; Sebastiano Salvidio; Slawomir Mitrus; Stefano Scali; Val- entin Pérez-Mellado; Vinicius Guerra; Wen Bo Liao; Xavier Bonnet; Xavier Santos. Acta Herpetologica 14(2): 71-80, 2019 DOI: 10.13128/a_h-7744
Podarcis siculus latastei (Bedriaga, 1879) of the Western Pontine Islands (Italy) raised to the species rank, and a brief taxonomic overview of Podarcis lizards
Gabriele Senczuk1,2,*, Riccardo Castiglia2, Wolfgang Böhme3, Claudia Corti1 1 Museo di Storia Naturale dell’Università di Firenze, Sede “La Specola”, Via Romana 17, 50125 Firenze, Italy. *Corresponding author. E-mail: [email protected] 2 Dipartimento di Biologia e Biotecnologie “Charles Darwin”, Università di Roma La Sapienza, via A. Borelli 50, 00161 Roma, Italy 3 Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, D53113, Bonn, Germany
Submitted on: 2019, 12th March; revised on: 2019, 29th August; accepted on: 2019, 20th September Editor: Aaron M. Bauer
Abstract. In recent years, great attention has been paid to many Podarcis species for which the observed intra-specific variability often revealed species complexes still characterized by an unresolved relationship. When compared to oth- er species, P. siculus underwent fewer revisions and the number of species hidden within this taxon may have been, therefore, underestimated. However, recent studies based on genetic and morphological data highlighted a marked differentiation of the populations inhabiting the Western Pontine Archipelago. In the present work we used published genetic data (three mitochondrial and three nuclear gene fragments) from 25 Podarcis species to provide a multilocus phylogeny of the genus in order to understand the degree of differentiation of the Western Pontine populations. In addition, we analyzed new morphometric traits (scale counts) of 151 specimens from the main islands of the Pontine Archipelago. The phylogenetic analysis revealed five principal Podarcis groups with biogeographic consistency. The genetic distinctiveness of the Podarcis populations of the Western Pontine Islands is similar or even more ancient than those observed in numerous other pairs of Podarcis sister species. In the light of these evidences we raise the Western Pontine lizards to specific rank; thus they should be referred to as Podarcis latastei.
Keywords. Reptilia, Podarcis latastei, Podarcis siculus, insular lizards, Mediterranean.
INTRODUCTION ing, as some predecessors did, only “varieties“ within this species (Boulenger, 1887, 1905, 1913, 1920). His main The wall lizards belonging to the genus Podarcis antagonist, representing the taxonomic “splitter“ fac- Wagler, 1830 are among the most speciose vertebrates in tion, was Ludwig von Méhely (1862-1953) who consid- Europe, representing one of the most important faunis- ered many of Boulenger’s “varieties” to be distinct species tic elements of the Mediterranean insular biota. Origi- (Méhely, 1907, 1909). He wrote: «Like a night-mare, the nally, two opposite taxonomic viewpoints (“lumping“ and so-called muralis question is burdening the mind of her- “splitting“) were – partly emotionally – discussed in the petologists» (Méhely, 1907). Despite modern approaches, late 19th and early 20th centuries. At that time the most molecular genetics included, Méhely was closer to the prominent representative of the “lumpers“ was George A. current concept than his more influential contempo- Boulenger (1859-1937) who joined numerous wall lizards rary colleague; however, the number of Podarcis species together under the name “Lacerta muralis“ distinguish- is still debated. For example, 21 taxa were recognized as
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 72 Gabriele Senczuk et alii valid species by Speybroeck et al. (2010), whereas other ventotenensis = Podarcis siculus ventotenensis (Taddei, authors have suggested 23 (Sindaco et al., 2013; Uetz and 1949) from Ventotene; Lacerta sicula pasquinii = Podar- Hošek, 2016; but see Psonis et al., 2017). The taxonomic cis siculus pasquinii (Lanza, 1952) from Scoglio Cappello wavering of the genus Podarcis is mainly due to the pres- near Palmarola; Lacerta sicula patrizii = Podarcis siculus ence of marked intra-specific variability with multiple patrizii (Lanza, 1952) from Zannone; Lacerta sicula lan- species complexes characterized by unresolved relation- zai = Podarcis siculus lanzai (Mertens, 1967) from Gavi ships (Harris and Arnold, 1999; Oliverio et al., 2000; and Lacerta sicula palmarolae = Podarcis siculus pal- Harris et al., 2005; Lymberakis et al., 2008). Table 1 sum- marolae (Mertens, 1967) from Palmarola (cfr. Lanza and marizes this taxonomic/nomenclatural history of the cur- Corti, 1996; Corti et al., 2010). rently recognized Podarcis species. Podarcis siculus parkeri was synonymized with P. s. In contrast to the great taxonomic attention paid sanctistephani (Mertens and Wermuth, 1960; Mertens, to numerous Podarcis species, P. siculus has undergone 1965), which is believed to have become extinct during fewer revisions and the number of species hidden within the first decades of the last century (1914 at the latest), this taxon may have been underestimated. Podarcis sicu- and replaced by P. s. siculus (Mertens, 1965). Henle and lus (Rafinesque-Schmaltz, 1810) was originally described Klaver (1986), reviewing the intraspecific taxa, followed as Lacerta sicula. However, because of its distribution Mertens (1965) in considering P. s. ventotenensis as a syn- over a large part of Italy (Sicily, Sardinia and numerous onym of the nominotypical form, and listed P. s. latastei, minor islands, islets and rocks) and Dalmatia, several P. s. lanzai, P. s. pasquinii, P. s. patrizii and P. s. pal- subspecies were described. Some of them were originally marolae as valid subspecies. These five taxa occur in the assigned to “Lacerta” muralis (more than 90 were listed Western Pontine Islands, which are believed to have been together with their type localities by Henle and Klaver, connected to the mainland in the Pleistocene, whereas 1986). This situation led some authors to hypothesize the the Eastern Pontine Islands (Ventotene, Santo Stefano) presence of a species complex similar to those observed seem never to have been, being located along the 500 m in other Podarcis assemblages (Oliverio et al., 1998, 2000; isobath (Woldstedt, 1958; Mertens, 1965, 1967). Harris and Sa-Sousa, 2002). More recent studies based on The deep genetic distance recently found between mitochondrial (Podnar et al., 2005) and nuclear (Senc- the Eastern and the Western Pontine Islands populations zuk et al., 2017) markers have supported the monophyly (Senczuk et al., 2018a), geometric morphometrics (Senc- of P. siculus and revealed surprisingly high genetic diver- zuk et al., 2018b), classical morphometric and meristic gences between the main constituent evolutionary line- data, as well as an updated time calibrated multilocus ages, most comparable to those observed between many phylogeny of Podarcis (Wagler, 1830), all suggest that the recognized Podarcis species (Harris et al., 2005). In addi- Western Pontine lizards deserve their own specific status tion, recent studies using molecular markers (mitochon- and should be referred to as Podarcis latastei (Bedriaga, drial and nuclear DNA) and geometric morphometrics 1879), which we characterize and redescribe here. have revealed that the populations from the Western Pontine Islands represent an evolutionarily independent unit (Senczuk et al., 2018a, 2018b). The genetic distances MATERIALS AND METHODS of these populations with respect to mainland ones were extraordinary high (p-distances of 7-10% for the mtDNA Molecular phylogenetics cytb gene), and the head morphology was clearly distin- guishable with respect to the mainland and Sicilian popu- To obtain a robust and time calibrated phylogeny of Podar- cis as a whole, three mitochondrial (16s; cytb, and nd4) and lations (Senczuk et al., 2018a; 2018b). three nuclear (mc1r, pod15b and pod55) gene fragments from The Pontine Archipelago is located 40 km off the 25 Podarcis species, including several subspecies, were retrieved Tyrrhenian coast and comprises the Western Pontine from GenBank (all samples are reported in Fig. 1 and Table islands Ponza, Palmarola, Zannone and Gavi, and the 1, localities and accession numbers are reported in Appendix Eastern Pontine islands Ventotene and Santo Stefano. 1, Table A1). Most of the retrieved sequences for each species From the Pontine Archipelago, the following nominal belong to the same individual, when not possible we selected insular intraspecific taxa have been described: Lacerta individuals of close geographic origin. All final consensus align- muralis var. latastei = Podarcis siculus latastei (Bedriaga, ments were computed for each gene separately using BioEdit 7.2 1879 a, b) from Ponza; Lacerta muralis parkeri = Podarcis (Hall, 1999). Coding gene fragments (cytb, nd4 and mc1r) were translated into amino acids to assess the lack of stop codons. siculus parkeri (Mertens, 1926) from Santo Stefano; Lac- For each alignment we used jModelTest v.2.1.3 (Darriba erta sicula sancti-stephani = Podarcis siculus sanctistepha- et al., 2012) to assess the best model of nucleotide evolution ni (Mertens, 1926) from Santo Stefano; Lacerta sicula under the corrected Akaike Information Criterion (AICc). To New Podarcis species from Western Pontine Islands 73
Fig. 1. Distribution of the genus Podarcis and location of the samples used for the phylogenetic analysis, as reported in Table 1. Geographic distribution of Podarcis latastei is also reported at the top right. reconstruct phylogenetic relationships we used a Bayesian coa- lus recognized by him and compared them with our own data lescent framework implemented in BEAST v.1.8 (Drummond taken from the holdings deposited in the Florence Museum et al., 2012). To calibrate the tree in absolute time we used two (MZUF). We measured and counted the scales of 151 speci- vicariant calibration points: the separation between the Pelo- mens (60 females and 91 males) from the main islands of the ponnesus (P. peloponnesiacus) and the islands of Crete and Pori Pontine Archipelago preserved at the Natural History Museum (P. cretensis and P. levendis); and the separation between the of the University of Florence (MZUF) (see Table 2). Speci- islands of Menorca and Mallorca (P. lilfordi) and the Pytiusic mens previously studied by Lanza (1952, 1967) and used for his Islands (P. pytiusensis). Both episodes occurred following the descriptions of P. s. patrizii and P. s. pasquinii were also includ- Messinian Salinity Crisis (MSC, at about 5.2 Mya) yielding the ed. We analyzed sex, snout-vent length (SVL), and the follow- sudden separation of these landmasses (Poulakakis et al., 2003; ing meristic characters: a) number of mid-body dorsal scales Brown et al., 2008). A normal distribution using the mean in (DORS); b) number of ventral plates counted longitudinally real space option (μ = 5.325; SD = 0.2) has been incorporated along the intermediate row (VENT); c) number of gular scales for each of the aforementioned nodes. We used a Yule process counted along the throat mid-line from the collar to the conflu- in a linked tree partition and a lognormal relaxed model main- ence of maxillaries (GUL); d) number of collar scales (COLL); taining unlinked clock partitions. As substitution models we e), number of femoral pores on the right leg (PORF). used GTR+I+G for 16s and cytb; TVM+I+G for nd4 and mc1r; To test for significance of differences between sexes and HKY for pod15b and HKY+I for pod55. We performed three islands, we used a two-way analysis of variance (ANOVA). An independent runs of 108 generations sampling every 104 steps. additional two-way ANOVA was performed to test differences Convergence was checked using the software TRACER v 1.5 between Ventotene Island and Santo Stefano Island sampled (Rambaut and Drummond, 2007) and after combining the trees in 1954 and 1966, and the Western Pontine and Santo Stefano using LogCombiner, the final consensus tree was computed in Island sampled in 1878. TreeAnnotator (Drummond et al., 2012).
RESULTS AND DISCUSSION Morphology The final alignment of 3117 bp included 27 taxa (Sup- We used the measurements and scale counts published by Mertens (1967) for diagnosing the subspecific taxa of P. sicu- plementary Information). The three independent runs 74 Gabriele Senczuk et alii
Table 1. Currently accepted Podarcis species and their original description name and reference. The geographic localities are shown in Fig. 1.
Loc. Species (Author and year of description) Described as Reference 1 P. hispanicus (Steindachner, 1870) Lacerta oxycephala var. hispanica Geniez et al. 2007 2 P. carbonelli Pérez-Mellado, 1981 Podarcis bocagei carbonelli Sá-Sousa and Harris, 2002 3 P. bocagei (Lopez-Seoane, 1885) Lacerta murals bocagei Sá-Sousa et al., 2000 Lacerta muralis var. liolepis 5 P. liolepis (Boulenger, 1905) Geniez et al., 2014 Lacerta muralis atrata 6 P. vaucheri (Boulenger, 1905) P. hispanicus vaucheri Oliverio et al., 2000 Lacerta muralis guadarramae, 7a, 7b P. guadarramae (Boscá, 1916) Geniez et al., 2014 Podarcis hispanicus “type 1A, 1B” 8 P. virescens (Geniez et al., 2014) Podarcis hispanicus “type 2” Geniez et al., 2014 9 P. muralis (Laurenti, 1768) Lacerta muralis 10 P. lilfordi (Günther, 1874) 11 P. pityusensis (Boscá, 1883) Lacerta muralis var. pityusensis 12 P. tiliguerta (Gmelin, 1789) Lacerta tiliguerta 13a, 13b P. siculus (Rafinesque-Schmaltz, 1810) Lacerta sicula 14 P. latastei (Bedriaga, 1876) 15 P. waglerianus (Gistel, 1868) Podarcis muralis var. wagleriana 16 P. raffoneae (Mertens, 1952) Lacerta sicula raffonei Capula, 1994 17 P. filfolensis (Bedriaga, 1876) 18 P. melisellensis (Braun, 1877) 19a, 19b P. tauricus (Pallas, 1814) 20 P. gaigeae (Werner, 1930) Lacerta taurica gaigeae 21 P. milensis (Bedriaga, 1882) Lacerta muralis fusca var. milensis 22 P. peloponnesiacus (Bibron and Bory, 1833) 23 P. erhardii (Bedriaga, 1882) Lacerta muralisfusca var. erhardii 24 P. cretensis (Wettstein, 1952) Lacerta erhardii cretensis 25 P. levendis (Lymberakis et al., 2008)
Table 2. Population number and relative sample size for both males responds to previous phylogenetic reconstructions. Within and females for each island. *Individuals collected in Santo Stefano the Podarcis radiation we found five principal groups with Island in 1954/1966. biogeographic consistency (Fig. 1-2). 1 – The Podarcis hispanicus complex currently N° Island Females Males includes seven species distributed from North Africa to 1 Ponza 18 22 the Iberian Peninsula and south-western France. All spe- 2 Gavi 3 9 cies from the P. hispanicus complex were first described 3 Palmarola 11 21 as intraspecific taxa of the collective species P. muralis 4 Zannone 6 6 and later raised to species rank in order to resolve para- 5 Santo Stefano (1878) 5 5 phyly (see Table 1) (Oliverio et al., 2000; Sá-Sousa and Santo Stefano* 5 14 Harris, 2002; Geniez et al., 2007, 2014). Our phyloge- 6 Ventotene 8 12 netic analysis support a similar phylogenetic relationships 7 Scoglio Cappello 4 2 among species as previously reported, and suggested, Tot. 60 91 albeit with moderate support (0.91), P. muralis as the sis- ter species of all the Iberian Podarcis. 2 – The “erhardii” group comprises species of the showed Effective Sample Size (ESS) for each parameter of Balkan Peninsula and the Greek islands. Because of a more than 200. The phylogenetic tree obtained is shown paraphyletic relationship between P. erhardii (Bedriaga, in Fig. 2. The tree topology is rather well supported (most 1882) and P. peloponnesiacus (Bibron and Bory, 1833), of the nodes showed posterior probabilities higher than two new insular endemics P. cretensis (Wettstein, 1952) 0.95) and the relationships among species only partly cor- and P. levendis (Lymberakis et al., 2008) were raised to New Podarcis species from Western Pontine Islands 75
Fig. 2. Bayesian phylogenetic tree based on multilocus data (cytb, 2003; Brown et al., 2008). The times of the most recent common 16s, nd4, mc1r, pod15b and pod55) using BEAST v. 1.8. Black filled ancestor are reported for each node as well as the posterior prob- circles indicate nodes used to calibrate phylogeny (Poulakakis et al., ability. the species rank (Poulakakis et al., 2003; 2005a; Lymbera- gaigeae, P. milensis, and six in the P. tauricus complex. kis et al., 2008). Based on the absence of support to the monophyly of P. 3 – The “tauricus” group includes two species P. tau - tauricus, the authors proposed to raise the subspecies P. ricus (Pallas, 1814) and P. melisellensis (Braun, 1877) dis- t. ionicus (Lehrs, 1902) to the species rank (Psonis et al., tributed over a large part of the Balkans and two endemic 2017). Our phylogenetic analysis confirms this scenario insular species: P. gaigae (Werner, 1930) from Skyros and indicating an ancient divergence between P. ionicus and P. surrounding islands, and P. milensis (Bedriaga, 1882) tauricus (Fig. 2). from Milos and surrounding islands (Poulakakis et al., It is interesting to note that although the geographic 2005a, 2005b). However, a recent species delimitation distribution of P. filfolensis (Bedriaga, 1876) would sug- approach (Psonis et al., 2017), suggested the presence of gest a close relationship with the other two endemic spe- nine species within the tauricus group: P. melisellensis, P. cies of the Siculo-Maltese area, P. waglerianus (Gistel, 76 Gabriele Senczuk et alii
Table 3. Scale counts after Mertens (1965) (minimum – mean – maximum) of the insular populations of P. latastei (the six left columns) and P. siculus (the two right columns). DORS = no. of mid-body dorsal scales; VENT = no. of ventral plates; COLL = no. of collar scales; PORF = no. of femoral pores on the right leg; m. = males; f. = females.
Ponza Gavi Zannone Palmarola Sc. Cappello S. Stefano 1878 S. Stefano 1963 Ventotene DORS m 68-70.4-75 71-76.3-81 66-72.8-78 69-76.6-86 72-73.2-76 72-75.8-79 60-65.6-72 61-66.8-78 f 62-67.7-73 70-73.2-78 63-68.0-74 66-68.7-71 66-68.7-71 71-75.3-79 59-61.2-63 60-60.3-68 VENT m 25-26.1-27 26-26.3-27 25-26.7-28 24-25.2-26 25-25.7-26 25-26.3-27 24-24.8-26 22-24.6-26 f 27-28.1-29 27-28.0-30 27-28.4-30 27-28.2-29 28-28.7-30 27-28.1-30 27-28.0-29 25-26.9-29 COLL. m 9-10.8-12 9-10.0-11 9-10.5-12 9-10.4-13 10-10.5-11 12-12.1-13 8-9.1-11 9-10.6-12 f 10-10.7-11 9-10.2-11 9-10.5-11 10-10.7-11 11-11.0-11 10-11.0-12 7-8.2-09 8-9.8-11 PORF. m 22-24.8-29 22-24.3-26 22-25.1-28 21-24.9-29 22-24.7-28 24-25.3-28 19-23.8-26 20-23.5-27 f 21-23.8-28 22-24.7-26 19-23.0-25 22-24.1-26 23-24.5-26 22-24.8-27 20-21.6-24 20-22.0-23
Table 4. Snout-vent length (SVL) and scale counts (minimum – mean – maximum) of specimens preserved at the Natural History Museum of the University of Florence (MZUF). DORS = no. of mid-body dorsal scales; VENT = no. of ventral plates; GUL = no. of gular scales; COLL = no. of collar scales; PORF = no. of femoral pores on the right leg; m. = males; f. = females. *Individuals collected in Santo Stefano Island in 1954/1966.
Ponza Gavi Zannone Palmarola Sc. Cappello S. Stefano 1878 S. Stefano* Ventotene SVL m. 58-68.6-78.8 70.5-73.2-78 70.4-67.1-76.2 58-67-75 62.5-65-67.5 69-76.1-81.5 67-73.5-81.6 60-70.7-77 f. 50-58.6-68.6 61-63.5-67 52.6-62.7-76.1 57.8-52-63 56.5-57.9-60 55-63.1-69 59-63.5-70 53.5-60.4-66 DORS m. 67-70.1-76 73-74.3-77 69-70.1-73 68-72.4-76 72 68-71-74 59-64.1-69 62-66-70 f. 62-68.2-74 67-72-75 63-67.3-74 63-68-74 65-67-69 69-71.8-75 55-58-63 57-61.4-67 VENT m. 18-19.9-21 22-23-24 19-21.3--20 18-19.3-21 20-20.5-21 20-20.3-21 17-18.9-21 17-19.2-20 f. 22-22.7-24 18-19.9-23 22-22.2-23 21- 22.4-24 24 19-21-23 20-21.4-23 21-21.6-23 COLL m. 11-12.7-14 11-12.3-13 10-12.1-13 10-12.4-15 12-13-14 13-14-15 9-12.3-16 10-12.6-15 f. 11-13-15 11-13-15 10-11.5-13 11-12.8-14 12-12.8-13 12-13.4-15 9-11.2-12 11-12.9-16 GUL m. 28-32.4--37 31-34.8-38 31-33.7-40 27-33--39 31 34-34.3-35 23-26.8-32 27-30.1-34 f. 27-31.2--35 32-33.3-35 31-33-36 27-33--36 23-24-25 33-36.4-40 23-25.4-28 25-27-29 PORF m. 22-24.5-26 23-24.7-26 23-25-26 21-24.8-28 25-25.5-26 26-26.5-27 21-22.2-24 21-23.7-27 f. 21-23.9-27 22-24.4-27 21-23.3-27 21-22.8-26 23-24-25 23-27.4-30 20-20.8-21 20-22-25
1868) and P. raffoneae (Mertens, 1952), previous molec- et al., 2008). The phylogenetic reconstruction reported ular analysis has resulted in contrasting phylogenies here, confirms the close relationship of these endemic regarding the position of these three species (Harris et Western Mediterranean species. al., 2005; Psonis et al., 2017; Salvi et al., 2017). Our phy- 5 – Podarcis species from the Italian Peninsula, Sic- logenetic reconstruction supports a tangled evolutionary ily and surrounding islands forms a monophyletic assem- history indicating P. filfolensis as the sister species of the blage that includes P. siculus, P. waglerianus and P. raf - Podarcis “tauricus” group (Fig. 2). foneae. The last of these was raised to the species rank 4 – Podarcis species from the Western Mediterrane- on the basis of allozyme analysis although further stud- an islands include P. tiliguerta (Gmelin, 1789), P. lilfordi ies showed relatively low genetic distances from P. w ag- (Günther, 1874) and P. pityusensis (Boscá, 1883). Podarcis lerianus (3.3% at cytochrome b), far lower than those tiliguerta distributed in Sardinia, Corsica and surround- observed between many other Podarcis species (Capula, ing islands, has also been argued to be a species complex 1994; Harris et al., 2005). Based on our data, the line- showing very deep phylogeographic discontinuities (Har- age including P. waglerianus and P. raffoneae is sister to ris et al., 2005; Rodriguez et al., 2017; Salvi et al., 2017; Podarcis siculus and the lizards of the Western Pontine Senczuk et al., 2019). On the other hand, P. lilfordi and P. Archipelago. The Western Pontine Podarcis are separat- pityusensis from the Balearic and Pityusic islands showed ed from P. siculus by approximately 4 Mya based on our closer phylogenetic relationship as a consequence of results. The genetic distinctiveness of these insular popu- vicariance following the Messinian Salinity Crisis (Brown lations is comparable or even greater than several other New Podarcis species from Western Pontine Islands 77
Table 5. Analysis of variance (ANOVA) for SVL and meristic charac- ters of the MZUF specimens. Significant p-value at 0.05 are marked in bold. Degrees of freedom (d.f.) are also reported. In the last col- umn, ANOVA results using the endemic insular taxon (P. latastei + the extinct P. s. sanctistephani) and introduced P. siculus as factors, are reported. SVL = snout-to-vent length; DORS = no. of mid-body dorsal scales; VENT = no. of ventral plates; GUL = no. of gular scales; COLL = no. of collar scales; PORF = no. of femoral pores. *Individuals collected in Santo Stefano Island in 1954/1966.
S. Stefano 1878 + W. Pontine/ SEX Islands S. Stefano* + Ventotene d.f. 1 7 2 SVL F 152.96 8.73 4.54 p <0.001 <0.001 <0.05 DORS. F 40.73 27.15 57.42 p <0.001 <0.001 <0.001 Fig. 3. Detail of the entry of a P. latastei specimen from Ponza col- lected by J. v. Bedriaga, in the catalogue of the Göttingen Zoological VENT. F 259.4 9.88 5.4 Museum. p <0.001 <0.001 <0.01 COLL. F 1.74 2.9 4.44 p 0.18 <0.01 <0.05 species, following Bedriaga, (1979b). GUL. F 5.71 24.86 62.13 In accepting the specific status for the lizards of the p <0.05 <0.001 <0.001 Western Pontine Islands, under the oldest name avail- PORF. F 15.41 10.71 57.42 able Podarcis latastei (Bedriaga, 1879), type locality Ponza p <0.001 <0.001 <0.001 Island, we nevertheless accept the infraspecific subdivi- sions within the Western Pontine Islands assigned by earlier authors to P. siculus. This means that the former pairs of Podarcis sister species (i.e., P. bocagei/P. g u ad ar- subtaxa of the latter taxon, viz. patrizii, pasquinii, lan- ramae, P. carbonelli/P. virescens, P. cretensis/P. levendis, P. zai etc. now become subspecies of P. latastei. The vari- gaigeae/P. milensis, P. tauricus/P. ionicus). ous island populations of Podarcis latastei in the Western Our morphological analysis substantially confirms Pontine Archipelago exhibit variable color patterns. The what Bedriaga (1879a) and Mertens (1965) already patterned color morphs often show a tendency for longi- observed. Indeed, we found significant differences com- tudinal stripes to dissolve into oblique bands, thus form- paring the specimens of the Western Pontine Islands and ing a reticulate pattern with light ocelli included (Fig. 4, 5 the Santo Stefano Island collected in 1878, with those and 6). In other individuals, particularly from Gavi Island collected in Santo Stefano in 1954/1966 and Ventotene (Fig. 5a, 5b) there is a strong tendency for a reduction Island (Tables 3, 4 and 5; Fig. A1). Furthermore, we also of black-pigmented color pattern elements, correspond- found significant differences when considering sexes and ing to the “concolor” mutation that also occurs in other islands as factors (Tables 3, 4 and 5; Fig. A1). Smaller Podarcis species. These differences in body dimensions, dorsal scales (resulting in higher dorsal scales counts) scalation and color pattern justify, in our opinion, the were already reported by Bedriaga (1879a) to characterize maintenance of their subspecific names, at least for con- his new taxon latastei. Slight discrepancies between the servation purposes (Senczuk et al., 2018a). scale counts taken by Mertens (1967) as compared with Bedriaga (1879a) based his nomen latastei on an ours (Tables 3 and 4) are likely due to a different count- unknown number of individuals – “in Anzahl” which ing method, which was not precisely defined in Mertens’ means “in a certain quantity” – collected by himself on (1967) paper, e.g., in the number of oblique ventral rows Ponza Island in Summer 1878, plus one individual from which is dependent on whether only complete or also a rock west off Ponza which he called Faraglioni of Ponza. incomplete rows are counted. Obviously, he kept all specimens in a cage alive during his Based on multiple sources of evidence from genetics travel and brought them via Nice (Nizza), France, from (herein and Senczuk et al., 2018a), morphology (herein where he sent a part of them to F. Lataste to Paris, to his and in Senczuk et al., 2018b) we believe that this insu- residential town of Heidelberg, Germany, where he con- lar endemic taxon deserves specific rank and should be tinued to observe them in life, mainly in respect to colour referred to as Podarcis latastei (Bedriaga, 1879). We pro- change phenomena (Bedriaga, 1879a). In 1879 and 1902 pose to adopt as common name “Lataste’s lizard” for this he sent preserved specimens to some German museums, 78 Gabriele Senczuk et alii
Fig. 5. Examples of living representatives of the Pontine Islands populations. Gavi: a, b, c; Ponza: d, e; Palmarola: f.
Fig. 4. Detail of plate IX. (Bedriaga, 1879b) with a specimen (right) of his “Lacerta muralis var. latastei (= Podarcis latastei). including Frankfurt and Munich as well as the Zoological Museum of the University of Göttingen (whose herpeto- logical holdings have been in Bonn since 1977), and one specimen of his Ponza lizard is still documented in the old Göttingen catalogue, although unfortunately it was lost some time before 1968 (Böhme, 2014, Fig. 3). We failed to retrieve any of these old syntypes in any of the mentioned collections. So, there seems to be no extant type mate- rial of this taxon, and the single colour image provided by Bedriaga (1879b), can be regarded as the figure of the individual that could have been chosen as a lectotype if it would be still extant (Fig. 4). A neotype selection, howev- er, seems presently to be unnecessary in this case. According to the genetic and geometric morphometric data published by Senczuk et al. (2018a, b) and to our data, the wall lizards of the Western Pontine Islands, so far clas- sified as belonging to Podarcis siculus, clearly merit their own specific status und should be treated under the oldest Fig. 6. Two different color morphs from Zannone Island: a “quasi” concolor individual above and a dark reticulated individual below. New Podarcis species from Western Pontine Islands 79 available name for these Western Pontine populations, i.e., Boulenger, G.A. (1920): Monograph of the Lacertidae. Podarcis latastei (Bedriaga, 1879). Because of the marked Vol. 1. Trustees of the British Museum, London. morphological differences between these populations, Brown, R.P., Terrasa, B., Pérez-Mellado, V., Castro, J.A., their former insular subspecific names (Ponza: latastei, Hoskisson, P.A., Picornell, A., Ramon, M.M. (2008). Gavi: lanzai, Zannone: patrizii, Palmarola: palmarolae, and Bayesian estimation of post-Messinian divergence Scoglio Cappello: pasquinii) which were ranked as sub- times in Balearic Island lizards. Mol. Phylogenet. Evol. species of P. siculus before, should be maintained but now 48: 350-358. attached as subspecific names to Podarcis latastei. Each of Capula, M. (1994): Genetic variation and differentiation these island populations has its own characteristics and in the lizard, Podarcis wagleriana (Reptilia: Lacerti- may well turn out to be a distinct conservation unit. dae). Biol. J. Linn. Soc. 52: 177-196. Corti C., Biaggini M., Capula M. (2010): Podarcis siculus (Rafinesque-Schmaltz, 1810). In: Fauna d’Italia. Reptilia, ACKNOWLEDGEMENTS pp. 407-417. Corti C., Capula M., Luiselli L., Razzetti E., Sindaco R., Eds, Edizioni Calderini, Bologna, Italy. We wish to thank Annamaria Nistri for allowing us Darriba, D., Taboada, G.L., Doallo, R., Posada, D. (2012): to access to the MZUF specimens, Ulla Bott for providing jModelTest 2: more models, new heuristics and paral- copies of some old papers and Jean-Michel Delaugerre for lel computing. Nat. Methods 9: 772-772. having provided the beautiful colour picture for the cover Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A. of Acta Herpetologiga. We also would like to thank Pao- (2012): Bayesian phylogenetics with BEAUti and the lo Colangelo for his helpful suggestions. All animals have BEAST 1.7. Molecular Biology and Evolution 29: been handled in accordance with relevant guidelines in full 1969-1973. compliance with specific permits released by the Italian Geniez, P., Cluchier, A., Sa-Sousa, P., Guillaume, C.P., Environment Ministry (Prot. 00017879/PNM-09/09/2012) Crochet, P.A. (2007): Systematics of the Podarcis his- and no lizards were killed. We thank the Circeo National panicus complex (Sauria,Lacertidae) I: Redefinition, Park for the permission to collect tissue samples from the morphology and distribution of the nominotypical Zannone Island population (N.487, 16/02/2015). taxon. Herpetol. J. 17: 69-80. Geniez, P., Sa-Sousa, P., Guillaume, C.P., Cluchier, A., Crochet, P.A. (2014): Systematics of the Podarcis his- SUPPLEMENTARY MATERIAL panicus complex (Sauria, Lacertidae) III: valid nomi- na of the western and central Iberian forms. Zootaxa Supplementary material associated with this article 3794: 1-51. can be found at
Lanza, B. (1967): Su due nuove razze insulari di Lacerta sic- locus phylogeny and species delimitation. Mol. Phylo- ula e di Lacerta tiliguerta. Arch. Zool. Ital. 51: 511-522. genet. Evol. 106: 6-17. Lanza, B., Corti, C. (1996). Evolution of the knowledge Rambaut, A., Suchard, M., Xie, D., Drummond, A. on the Italian herpetofauna during the 20th century. (2014): Tracer v1. 6. http://beast.bio.ed.ac.uk/Tracer/ Boll. Mus. civ. St. Nat. Verona 20: 373-436. [accessed on 29 May 2015] Lymberakis, P., Poulakakis, N., Kaliontzopoulou, A., Rodríguez, V., Buades, J.M., Brown, R.P., Terrasa, B., Valakos, E., Mylonas, M. (2008): Two new species of Pérez-Mellado, V., Corti, C., Delaugerre, M., Castro, Podarcis (Squamata; Lacertidae) from Greece. Syst. J. A., Picornell, A., Ramon, C. (2017): Evolutionary Biodiv. 6: 307-318. history of Podarcis tiliguerta on Corsica and Sardinia. Méhely, L. v. (1907): Zur Lösung der „Muralis-Frage“. BMC Evol. Biol., 17: 27. Vorläufige Mitteilung. Ann. Mus. Natl. Hungar. 5: Salvi, D., Pinho, C., Harris, D.J. (2017): Hotspot of genet- 84-88. ic diversity: decoupled mito-nuclear histories in the Méhely, L. v. (1909): Materialien zu einer Systematik und evolution of the Corsican-Sardinian endemic lizard Phylogenie der muralis-ähnlichen Lacerten. Ann. Podarcis tiliguerta. BMC Evol. Biol. 17: 63. hist.-nat. Mus. Natl. Hungar. 7: 409-621. Sà-Sousa, P., Harris, D.J. (2002): Podarcis carbonelli Pérez- Mertens, R. (1926): Zwei neue Inselrassen der Gattung Mellado, 1981 is a distinct species. Amphibia-Reptilia Lacerta. Zool. Anz. 68: 319-322. 23: 459-468. Mertens, R. (1965): Das Rätsel der Eidechsen von Santo Senczuk, G., Colangelo, P., De Simone, E., Aloise, G., Stefano. Zool. Jahrb., Abt, Syst. 92: 91-102. Castiglia, R. (2017): A combination of long term frag- Mertens, R. (1967): Unterlagen zu einer „Herpetolo- mentation and glacial persistence drove the evolution- gia tyrrhenica“ VIII. Die Reptilien der Pontinischen ary history of the Italian wall lizard Podarcis siculus. Inseln. Senckenberg. biol. 44: 125-144. BMC Evol. Biol. 17: 6. Mertens, R., Wermuth, H. (1960): Die Amphibien und Senczuk, G., Havenstein, K., Milana, V., Ripa, C., De Reptilien Europas (Dritte Liste nach dem Stand von Simone, E., Tiedemann, R., Castiglia, R. (2018a): Spot- vom 1. Januar 1960). Kramer, Frankfurt am Main. light on islands: on the origin and diversification of an Oliverio, M., Bologna, M.A., Mariottini, P. (2000): Molec- ancient lineage of the Italian wall lizard Podarcis sicu- ular biogeography of the Mediterranean lizards Podar- lus in the western Pontine Islands. Sci. Rept. 8: 15111. cis Wagler, 1830 and Teira Gray, 1838 (Reptilia, Lacer- Senczuk, G., Colangelo, P., Avramo, V., Castiglia, R. tidae). J. Biogeogr. 27: 1403-1420. Böhme, W., Corti, C. (2018b): A study in scarlet: Podnar, M., Mayer, W., Tvrtković, N. (2005): Phyloge- incipient speciation, phenotypic differentiation and ography of the Italian wall lizard, Podarcis sicula, as conservation implications of the Podarcis lizards of revealed by mitochondrial DNA sequences. Mol. Ecol. the western Pontine Islands, Italy. Biol. J. Linn. Soc. 14: 575-588. 125: 50-60. Poulakakis, N., Lymberakis, P., Antoniou, A., Chalkia, D., Senczuk, G., Castiglia, R., Colangelo, P., Delaugerre, M., Zouros, E., Mylonas, M., Valakos E. (2003): Molecular Corti, C. (2019): The role of island physiography in phylogeny and biogeography of the wall-lizard Podar- maintaining genetic diversity in the endemic Tyrrhe- cis erhardii (Squamata: Lacertidae). Mol. Phylogenet. nian wall lizard (Podarcis tiliguerta). J. Zool. 309: 140- Evol. 28: 38-46. 151. Poulakakis, N., Lymberakis, P., Valakos, E., Pafilis, P., Sindaco, R, Venchi, A, Grieco, C. (2013): The reptiles of Zouros, E., Mylonas, M. (2005): Phylogeography of the western Palearctic, Volume 2 Annotated checklist Balkan wall lizard (Podarcis taurica) and its relatives and distributional atlas of the snakes of Europe, North inferred from mitochondrial DNA sequences. Mol. Africa, Middle East and Central Asia, with an update Ecol. 14: 2433-2443. to Volume 1. Edizioni Belvedere, Latina, Italy. Poulakakis, N., Lymberakis, P., Valakos, E., Zouros, E., Speybrock, J., Beukema, W., Crochet, P.-A. (2010): A tenta- Mylonas, M. (2005b): Phylogenetic relationships and tive species list of the European herpetofauna (Amphib- biogeography of Podarcis species from the Balkan ia and Reptilia) – an update. Zootaxa 2492: 1-27. Peninsula, by Bayesian and maximum likelihood anal- Taddei, A. (1949): Le Lacerte (Archaeolacerta e Podar- yses of mitochondrial DNA sequences. Mol. Phylo- cis) dell’Italia peninsulare e delle isole. Coment. Pont. genet. Evol. 37: 845-857. Acad. Sci. Vaticano 13: 197-174. Psonis, N., Antoniou, A., Kukushkin, O., Jablonski, D., Woldstedt, P. (1958): Das Eiszeitalter. Grundlinien einer Petrov, B., Crnobrnja-Isailović, J., Poulakakis, N. Geologie des Quartärs, II. Europa, Vorderasien und (2017): Hidden diversity in the Podarcis tauricus (Sau- Nordafrika im Eiszeitalter, 2nd Edition. Enke, Stutt- ria, Lacertidae) species subgroup in the light of multi- gart, Germany. Acta Herpetologica 14(2): 81-87, 2019 DOI: 10.13128/a_h-7745
Substrate type has a limited impact on the sprint performance of a Mediterranean lizard
Pantelis Savvides1,*, Eleni Georgiou1, Panayiotis Pafilis2,3, Spyros Sfenthourakis1 1 Dept. of Biological Sciences, University of Cyprus, Nicosia, Cyprus. *Corresponding author. Email: [email protected] 2 Section of Zoology and Marine Biology, Dept. of Biology, National and Kapodistrian University of Athens, Greece 3 Zoological Museum, National and Kapodistrian University of Athens, Greece
Submitted on: 2018, 7th November; revised on: 2019, 28th February; accepted on: 2019, 29th May Editor: Simon Baeckens
Abstract. Environmental factors may affect animal performance in diverse ways, even among different populations of a single species. Here, we assess the impact of substrate type on the sprint performance (maximum speed and acceler- ation) of Schreiber’s fringe-fingered lizard (Acanthodactylus schreiberi). This species is a skillful runner that also bears micro spike-like protruding scales on its toepads (toe fringes), an adaptation for locomotion on sand. We worked with three populations living in habitats that differ in substrate type (sand, soil and rock). We measured sprint perfor- mance using a race-track with custom substrate platforms replicating the different substrate types. We formulated two hypotheses: first, we anticipated that the three populations would differ in their sprint performance due to the differ- ences in substrate type; second, we expected that each population would perform better on its home substrate. Our results generally refuted the hypothesis that sprint performance would differ on different substrate types. Our results suggest that there is a restricted effect of substrate type on locomotion and indicate a multifactor interplay among alternative underlying parameters.
Keywords. Ecophysiology, morphology, locomotion, Lacertidae, Cyprus.
INTRODUCTION because of their distinct ecology (Aerts et al., 2000). Also, Losos (1990) reported that locomotion parameters and Sprint performance is very important for all animals, morphological features evolved concordantly in 15 Anolis as it affects most of their daily activities. Sprinting is quite species. common among lizards during foraging, antipredator The results of previous studies on the effects of defense and inter- and intraspecific competitive behavior substrate type on locomotion are puzzling (Korff and (Losos and Irschick, 1996; Husak et al., 2006; McElroy et McHenry, 2011; Tulli et al., 2012; Vanhooydonck et al., al., 2008). Speed and acceleration, the main components 2015). The texture complexity (e.g., particle size, shape of sprint performance, may be crucial for the overall fit- and roughness) of the substrate type is also known to ness of individuals (Jayne and Bennett, 1990; Robson affect locomotor performance (Brandt et al., 2015; Berg- and Miles, 2000; Miles, 2004). Interactions between the mann et al., 2017). The propulsive forces applied by the ecology and morphology of species act as driving factors toes on non-solid substrates (e.g., sand) may be reduced that exert strong selective pressures leading to optimal because of insufficient grip and friction with the sub- locomotor performance (Van Damme et al., 2003; Husak strate, therefore leading to suboptimal sprint perfor- et al., 2006). For instance, gekkonid and lacertid liz- mance (Redfern et al., 2001; Korff and McHenry, 2011; ards were shown to adopt different locomotion patterns Brandt et al., 2015). Stiff, rough substrates provide more
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 82 Pantelis Savvides et alii friction that enhances grip, thus allowing higher per- formance (Kerdok et al., 2002; Van der Tol et al., 2005; Brandt et al., 2015; Bergmann et al., 2017). Morphological features that are often beneficial on certain substrate types include adaptations of the epi- dermis that covers the digits. For instance, the adhesive ability of many Gekkonidae species depends on toe pad microarchitecture (setae) that allows them to run eas- ily on smooth and vertical surfaces such as walls or even glass (Autumn et al., 2002; Autumn et al., 2005). Also, some other taxa bear fringes on their toes (e.g. genera Acanthodactylus, Basiliscus and Uma) enabling them to run fast on non-solid substrates (e.g., sand or water) without ‘sinking’, as fringes increase the amount of toe surface that comes into contact with the substrate while Fig. 1. The three habitats studied. A. Geri, B. Agros, C. Akrotiri and running (Salvador, 1982; Luke, 1986; Carothers, 1986). the respective platforms that were used in the laboratory to simu- In this study, we aimed to evaluate the impact of sub- late the substrate of each habitat. strate type on sprint performance (maximum speed and maximum instant acceleration) in Schreiber’s fringe-fin- We captured a total of 67 adult individuals of both sex- gered lizard (Acanthodactylus schreiberi Boulenger, 1878). es (excluding gravid females) (Geri, N = 22; Agros, N = 22; We worked with three Cypriot populations that reside Akrotiri, N = 23). Lizards were housed in individual terraria (30 in habitats with different substrate types (soil free from x 30 x 30 cm) in the laboratory under a constant temperature rocks, rock and sand). We formulated two hypotheses. (30 oC) and controlled photoperiod (16-h light and 8-h dark), First, we anticipated that the sprint performance of the and were provided with mealworms (Tenebrio molitor larvae) focal populations would differ due to the presence of dif- and fresh water ad libitum. All lizards were released at their ferent substrate types in the habitats and due to the dif- sampling sites after the completion of the experiments (experi- mental lizards remained in the laboratory for two weeks). ferent running styles required on each one (Van Damme Based on the substrate type of each habitat, we construct- et al., 1998). Second, we predicted that individuals would ed three removable substrate platforms that were used on a perform better on their home substrates than those com- custom-made wooden racetrack (240 x 12 cm2), bearing 10 cm ing from other habitats (Goodman et al., 2008). increments on its back and clear acrylic glass on its front in order to allow video recording of each trial (Fig. 1).
MATERIALS AND METHODS Morphological measurements Study system Snout-vent length (SVL), hind limb length (HLL) and hind Acanthodactylus schreiberi is a medium-sized lacertid liz- toe length (HT) were recorded with a digital caliper (Silverline ard (snout-vent length 73-93 mm for males and 55-76 mm for 380244, accurate to 0.01 mm), before running trials. Also, the females), inhabiting various habitats all over Cyprus (Baier et length of the largest right hind toe, the total number of fring- al., 2009). Even though it is considered to be mostly a sand- es on it and the length of the three largest fringes from base to dwelling lizard, it can be found from coastal areas to mountain tip were measured using stereoscopic images of their toes (N pine forests (over 1,300 m a.s.l) (Baier et al., 2009). The species = 60, 10 males and 10 females from each population) (Fig. 2), is a skillful and swift runner that can use bipedalism while run- in order to test for correlation between toe length and fringe ning (Savvides et al., 2017). microarchitecture (length and number of fringes). The habitats of the three focal populations vary consider- ably in substrate and vegetation type (Fig. 1). Geri (35o05’50”N, 33o26’21”E, elevation 183 m a.s.l.) is a sub-urban shrubland Sprint performance characterized by the presence of boxthorn (Lycium ferocissi- mum), thorny burnet (Sarcopoterium spinosum) and conehead All individuals were allowed to thermoregulate for an hour thyme (Thymbra capitata). The substrate consists mostly of solid in a specifically designed terrarium (Van Damme et al., 1986) soil without any rocks. Agros (34o56’27”N, 33o00’14”E, eleva- before each trial, so as to perform at the highest level possi- tion 1,348 m a.s.l.) is in a pine forest (Pinus brutia) with dense ble (Irschick and Losos, 1998). After this period, lizards were shrubs and has a rocky soil as its substrate. Akrotiri (34o56’27”N, placed in the racetrack. We triggered motion with a brush 33o00’14”E, elevation 1 m a.s.l.) is a coastal dune habitat with touching the lizard’s tail base. Each individual performed five quite sparse phrygana, where the substrate is fine-grained sand. trials on each substrate type in a single day. Between trials on Substrate type and speed performance in a lizard 83
tests. Log-transformed data for the morphological characters were compared between sexes for each population using one- way MANCOVA and taking SVL as covariate. One-way MAN- COVA was also used to search for differences among popula- tions in relation to the transformed data for the morphologi- cal characters (HLL and HT), using again SVL as a covariate. One-way MANOVA was used to compare their sprint per- formance among different substrate types and populations. A post-hoc Tukey test was used in order to determine the differ- ences among populations. The Friedmann test and the Bonfer- roni correction were used to compare the performance of each individual on the three types of substrate. The effects of the log- transformed values of morphological characters (HLL and HT) on sprint performance were identified independently, using lin- ear regression. Fig. 2. Stereoscopic image of the toes and an example of how we measured the three largest toe fringes. RESULTS different substrate types, lizards were left to rest for a day and Morphological characters showed significant differ- were then tested on the new type (in total, 15 trials per individ- ences between the sexes in each population, with female ual within five days). Trials were recorded with a video camera individuals having relatively smaller hind limbs and hind (Olympus SH-60) viewing the racetrack from the side (cover- toe length than male individuals (one-way MANCOVA: ing all the distance from start to finish), at a rate of 240 frames Geri, F2,18 = 9.965 Roy’s largest root = 1.107; Agros, F2,18 per second. Sprint performance was estimated from the video = 29.373 Roy’s largest root = 3.264; Akrotiri, F2,18 = 5.517 recordings (Martin and Avery, 1998; Kaliontzopoulou et al., Roy’s largest root = 0.613; all P values < 0.05). Thus, hind 2012; Vanhooydonck et al., 2015; Savvides et al., 2017). Maximum speed was calculated for all trials, based on the limbs and hind toes length were therefore compared number of frames needed to cover a distance of 20 cm within among populations separately for males and females a known time interval (Savvides et al., 2017). We chose the (Table 1), but no significant differences were found (all P highest values for each individual to represent its best trial. values > 0.05). Maximum instant acceleration was calculated by digitizing the When comparing sprint performance among popu- position of the lizard’s snout in every frame, for all trials, on lations, we observed that the Agros males (home habitat x- (movement) and y- (gravity) axes (MATLAB DLTdataview- with rocky substrate) were significantly faster than the er3; Hedrick, 2008). In each frame, we estimated the displace- other populations (one-way MANOVA: F6,26 = 4.072, ment of the snout and converted it from pixels to meters. We Roy’s largest root = 0.940, p = 0.005). They achieved the filtered the curve of the instantaneous displacement of the snout over time (i.e. instantaneous velocity) using a fourth-order zero highest maximum speed (Post hoc Tukey HSD: Agros vs. phase shift Butterworth low-pass data noise filter (40Hz; VBA Geri, P = 0.012; Agros vs. Akrotiri, P = 0.002) and maxi- application in Office Excel). The time differential of the instan- mum instant acceleration (Post hoc Tukey HSD: Agros taneous velocity yields the instantaneous acceleration and we vs. Geri, P = 0.001; Agros vs. Akrotiri, P = 0.002) on soil chose the highest value (i.e. peak) for each individual’s best trial (Van Wassenbergh, 2007). We only used trials during which liz- ards ran continuously for at least 50 cm. Using the Pearson product moment correlation coeffi- Table 1. Mean values for morphological characters between males cient, we found that the length of the largest hind toe correlated and females from the three populations. SVL: snout-vent length, strongly with the length of the three longest fringes in all popu- HLL: hind limb length, HT: longest hind toe length. Values are in cm. lations (all r values > 0.6 and P values < 0.05), so we used this measurement as a proxy of fringe size to detect possible fringe Geri Agros Akrotiri effects on sprint performance. Character Mean SD Mean SD Mean SD SVL ♂♂ 7.10 0.43 6.90 0.64 6.35 0.40 Statistical analyses ♀♀ 6.45 0.41 6.50 0.34 6.00 0.44 HLL ♂♂ 4.60 0.23 4.65 0.14 4.49 0.43 We used the Shapiro – Wilks and Levene’s tests to check ♀♀ 4.00 0.28 4.00 0.16 3.96 0.22 for data normality and homogeneity of variance, respectively. HT ♂♂ 1.29 0.14 1.24 0.12 1.22 0.10 All data were log-transformed based on the results of these ♀♀ 1.13 0.22 1.10 0.09 1.00 0.10 84 Pantelis Savvides et alii
Fig. 3. Mean values for speed and acceleration on each substrate within populations and for males and females.
(Fig. 3). We did not observe further differences in the favor higher performance (e.g., solid substrates like rock), sprint performance among other populations on the oth- while others restrict locomotion (e.g., substrates not pro- er substrates. viding sufficient grasp, e.g., sand or mud) (Vanhooydonck The sprint performance of all lizard populations (for et al., 2005; Tulli et al., 2012). In our study, we did indeed both sexes) did not differ among the three types of sub- see certain differences in sprint performance among con- strate (all P values > 0.05) (Fig. 3). specific populations of Schreiber’s fringe-fingered lizard. The length of hind limbs and toes showed no signifi- However, these differences followed a rather unclear pat- cant effects on performance among populations and sub- tern and indicated a limited effect of substrate type on strates (all regression P values > 0.05). locomotion. Interestingly, when we analyzed sprint performance taking into account the effect of hind limb length, we DISCUSSION did not find any significant effects for any type of sub- strate. Also, the multiple regression analyses did not show Locomotion patterns may change in response to a beneficial effect of toe fringes (using the hind toe as a endogenous or extraneous factors (Vanhooydonck and proxy). Hind limbs have been repeatedly reported to Van Damme, 2003; Sathe and Husak, 2018). Among the affect lizard locomotion (Vanhooydonck et al., 2001; Her- latter, the type of substrate is known to affect locomo- rel et al., 2002; Savvides et al., 2017). The absence of such tion in lizards. In some cases, specific substrate types effects in our study might indicate an interplay among Substrate type and speed performance in a lizard 85 other morphological features (e.g., fore limbs, tail length, on the fascinating interplays taking place in lizard loco- etc.) involved in locomotion in response to substrate type motion. requirements (Herrel et al., 2002). On the other hand, we cannot rule out the possibility that despite our best efforts, lizards might have underperformed and have ACKNOWLEDGMENTS failed to achieve their maximum performance levels. Contrary to our first hypothesis regarding sprint per- The study was carried out according to the Cyp- formance on different substrates, no differences emerged riot National Law on Animal Rights and Welfare (Law from the comparison among the three populations, save 55(I)/2013 for Animal Use on Scientific Experiments) the single case of the Agros males (rock substrate) that and under a permit issued by the Ministry of Agricul- were the fastest sprinters. However, subsequent analy- ture, Rural Development and Environment (Permit no: sis based on morphological features, failed to provide an 02.15.001.003, 04.05.002.005.006). We are grateful to Dr. underlying reason for this finding. In our initial hypoth- Anna-Nicola Chapman for her assistance. esis, we considered the three types of substrate to be of different quality. We presumed that sand would be the more challenging substrate because of its grainy texture REFERENCES that “sinks” under the weight of a running lizard (Clem- ente, 2014; Sathe and Husak, 2015). As such, we expected Aerts, P., Van Damme, R., Vanhooydonck, B., Zaaf, A., that sprint performance therein would be the lowest. On Herrel, A. (2000): Lizards locomotion: how morphol- the other hand, solid substrates are known to provide ogy meets ecology. Netherl. J. Zool. 50: 261-277. high traction and thus facilitate locomotion (Lejeune et Autumn, K., Buehler, M., Cutkosky, M., Fearing, R.S., al., 1998; Claussen et al., 2002; Brandt et al., 2015; Berg- Full, R.J., Goldman, D.I., Groff, R., Provancher, W., mann et al., 2017). Apparently, these predictions were not Rizzi, A.A., Saranli, U., Saunders, A., Koditschek, valid in our study system, as we failed to find any signifi- D.E. (2005): Robotics in scansorial environments. cant deviations. It seems that there is no ideal substrate In: Unmanned ground vehicle technology VII, pp. type for A. schreiberi locomotion and all populations are 291-303. Gerhart, G.R., Shoemaker, C.M., Gage, D.W., well adapted in their home habitat. Eds, SPIE Proceedings. According to our second hypothesis, lizards should Autumn, K., Sitti, M., Liang, Y.A., Peattie, A.M., Hans- have performed at higher levels on their home substrate. en, W.R., Sponberg, S., Kenny, T.W., Fearing, R., However, we did not find such a pattern in our study Israelachvili, J.N., Full, R.J. (2002): Evidence for van system. All populations performed at similar levels on der Waals adhesion in gecko setae. Proc. Natl. Acad. all substrate types, indicating that the species conserves Sci. 99: 12252-12256. generalized running capabilities that allow a high level of Baier, F., Sparrow, D.J., Wield, H.J. (2009): The amphib- performance on various types of substrate. The rejection ians and reptiles of Cyprus. Edition Chimaira, Frank- of the specialized populations hypothesis, might be due furt am Main. to overlapping genetic pools or mixed habitat character- Brandt, R., Galvani, F., Kohlsdorf, T. (2015): Sprint per- istics, leading the lizards in similar directions regarding formance of a generalist lizard running on different their kinematics and their interactions between morphol- substrates: grip matters. J. Zool. 297: 15-21. ogy, biomechanics and environmental factors. We also Bergmann, P.J., Pettinelli, K.J., Crockett, M.E., Schaper, have to highlight the fact that long periods of draught in E.G. (2017): It’s just sand between the toes: how par- Cyprus can change abruptly to rainy periods (especially ticle size and shape variation affect running perfor- during late spring and early summer). This would cause mance and kinematics in a generalist lizard. J. Exp. the substrate to change its properties, such as its rough- Biol. 220: 3706-3716. ness and its potential to provide grip and would thus Carothers, J.H. (1986): An experimental confirmation of favor the generalized pattern we observed in this study. morphological adaptation: toe fringes in the sand- This study will enhance the growing body of litera- dwelling lizard Uma scoparia. Evolution 40: 871-874. ture on saurian locomotion, as it examines some of the Claussen, D.L., Lim, R., Kurz, M., Wren, K. (2002): widely accepted principles in the field. According to our Effects of slope, substrate, and temperature on the findings, the type of substrate has a limited impact on locomotion of the ornate box turtle, Terrapene orna- sprint performance, and at least for Schreiber’s fringe- ta. Copeia 2002: 411-418. fingered lizard, there were no strict patterns observed. Clemente, C.J. (2014): The evolution of bipedal running Further research including more species will shed light in lizards suggests a consequential origin may be 86 Pantelis Savvides et alii
exploited in later lineages. Evolution 68: 2171-2183. Miles, D.B. (2004): The race goes to the swift: fitness con- Goodman, B.A., Miles, D.B., Schwarzkopf, L. (2008): Life sequences of variation in sprint performance in juve- on the rocks: habitat use drives morphological and nile lizards. Evol. Ecol. Res. 6: 63-75. performance evolution in lizards. Ecology. 89: 3462- Redfern, M.S., Cham, R., Gielo-Perczak, K., Grön- 3471. qvist, R., Hirvonen, M., Lanshammar, H., Marpet, Herrel, A., Meyers, J.J., Vanhooydonck, B. (2002): Rela- M., Pai, C.Y.C., Powers, C. (2001): Biomechanics of tions between microhabitat use and limb shape in slips. Ergonomics 44: 1138-1166. phrynosomatid lizards. Biol. J. Linn. Soc. 77: 149-163. Robson, M.A., Miles, D.B. (2000): Locomotor perfor- Husak, J.F., Fox, S.F., Lovern, M.B., Van Den Bussche, mance and dominance in male tree lizards, Urosaurus R.A. (2006): Faster lizards sire more offspring: sexual ornatus. Funct. Ecol. 14: 338-344.29. selection on whole-animal performance. Evolution 60: Salvador, A. (1982): A revision of the lizards of the genus 2122-2130. Acanthodactylus (Sauria: Lacertidae). Bonn. Zool. Irschick, D.J., Losos, J.B. (1998): A comparative analy- Monogr. 16: 1-167 sis of the ecological significance of maximal locomo- Sathe, E.A., Husak, J.F. (2015): Sprint sensitivity and loco- tor performance in Caribbean Anolis lizards. Evolu- motor trade-offs in green anole (Anolis carolinensis) tion 52: 219-226. lizards. J. Exp. Biol. 218: 2174-2179. Jayne, B.C., Bennett, A.F. (1990): Selection on locomotor Sathe, E.A., Husak, J.F. (2018): Substrate-specific loco- performance capacity in a natural population of garter motor performance is associated with habitat use in snakes. Evolution 44: 1204-12. six-lined racerunners (Aspidoscelis sexlineata). Biol. J. Kaliontzopoulou, A., Adams, D.C., van der Meijden, A., Linn. Soc. 124: 165-173. Perera, A., Carretero, M.A. (2012): Relationships Savvides, P., Stavrou, M., Pafilis, P., Sfenthourakis, S. between head morphology, bite performance and (2017): Tail autotomy affects bipedalism but not sprint ecology in two species of Podarcis wall lizards. Evol. performance in a cursorial Mediterranean lizard. Sci. Ecol. 26: 825-845. Nat. 104: 3. Kerdok, A.E., Biewener, A.A., McMahon, T.A., Weyand, Tulli, M.J., Abdala, V., Cruz, F.B. (2012): Effects of differ- P.G., Herr, H.M. (2002): Energetics and mechanics of ent substrates on the sprint performance of lizards. J. human running on surfaces of different stiffnesses. J. Exp. Biol. 215: 774-784. Appl. Physiol. 92: 469-478. Van Damme, R., Aerts, P., Vanhooydonck, B. (1998). Var- Korff, W.L., McHenry, M.J. (2011): Environmental differ- iation in morphology, gait characteristics and speed ences in substrate mechanics do not affect sprinting of locomotion in two populations of lizards. Biol. J. performance in sand lizards (Uma scoparia and Cal- Linn. Soc. 63: 409-427. lisaurus draconoides). J. Exp. Biol. 214: 122-130. Van Damme, R., Bauwens, D., Verheyen, R.F. (1986): Lejeune, T.M., Willems, P.A., Heglund, N.C. (1998): Selected body temperatures in the lizard Lacerta Mechanics and energetics of human locomotion on vivipara: variation within and between populations. J. sand. J. Exp. Biol. 201: 2071-2080. Therm. Biol. 11: 219-222. Lleonart, J., Salat, J., Torres, G.J. (2000): Removing allo- Van Damme, R., Vanhooydonck, B., Aerts, P., De Vree, F. metric effects of body size in morphological analy- (2003): Evolution of lizard locomotion: context and sis. J. Theor. Biol. 205: 85-93. constraint. In: Vertebrate biomechanics and evolu- Losos, J.B. (1990): Concordant evolution of locomotor tion, pp. 267-282. Bels, V., Gasc, J.P., Casinos, A., Eds, behaviour, display rate and morphology in Anolis liz- BIOS Scientific, Oxford. ards. Anim. Behav. 39: 879-890. Van der Tol, P.P.J., Metz, J.H.M., Noordhuizen-Stassen, Losos, J.B., Irschick, D.J. (1996): The effect of perch diam- E.N., Back, W., Braam, C.R., Weijs, W.A. (2005): Fric- eter on escape behaviour of Anolis lizards: laboratory tional forces required for unrestrained locomotion in predictions and field tests. Anim. Behav. 51: 593-602. dairy cattle. J. Dairy Sci. 88: 615-624. Luke, C. (1986): Convergent evolution of lizard toe fring- Vanhooydonck, B., Andronescu, A., Herrel, A., Irschick, es. Biol. J. Linn. Soc. 27: 1-16. D.J. (2005): Effects of substrate structure on speed and Martin, J., Avery, R.A. (1998): Effects of tail loss on the acceleration capacity in climbing geckos. Biol. J. Linn. movement patterns of the lizard, Psammodromus algi- Soc. 85: 385-393. rus. Funct. Ecol. 12: 794-802. Vanhooydonck, B., Measey, J., Edwards, S., Makhubo, McElroy, E.J., Hickey, K.L., Reilly, S.M. (2008): The cor- B., Tolley, K.A., Herrel, A. (2015): The effects of sub- related evolution of biomechanics, gait and foraging stratum on locomotor performance in lacertid liz- mode in lizards. J. Exp. Biol. 211: 1029-1040. ards. Biol. J. Linn. Soc. 115: 869-881. Substrate type and speed performance in a lizard 87
Vanhooydonck, B., Van Damme, R. (2003): Relationships between locomotor performance, microhabitat use and antipredator behaviour in lacertid lizards. Funct. Ecol. 17: 160-169. Vanhooydonck, B., Van Damme, R., Aerts, P. (2001): Speed and stamina trade-off in lacertid lizards. Evolu- tion 55: 1040-1048. Van Wassenbergh, S. (2007): https://www.uantwerpen.be/ en/staff/sam-vanwassenbergh/my-website/excel-vba- tool.
Acta Herpetologica 14(2): 89-100, 2019 DOI: 10.13128/a_h-7746
Coping with aliens: how a native gecko manages to persist on Mediterranean islands despite the Black rat?
Michel-Jean Delaugerre1,*, Roberto Sacchi2, Marta Biaggini3, Pietro Lo Cascio4, Ridha Ouni5, Claudia Corti3 1 Conservatoire du littoral, Résidence St Marc, 2, rue Juge Falcone F-20200 Bastia, France. *Corresponding author. Email: [email protected] 2 Dipartimento di Scienze della Terra e dell’Ambiente, Università degli Studi di Pavia, via Taramelli 24, I-27100, Pavia, Italia 3 Sistema Museale di Ateneo - Museo di Storia Naturale dell’Università di Firenze, Sede “La Specola”, Via Romana 17, I-50125 Firenze, Italia 4 Associazione Nesos, via Vittorio Emanuele 24, I-98055 Lipari (ME), Italia 5 Tunisia Wildlife Conservation Society, Faculté des Sciences de Tunis, Université Tunis El Manar II Tu-2092, Tunisia
Submitted on: 2019, 14th April; revised on: 2019, 16th June; accepted on: 2019, 24th June Editor: Uwe Fritz
Abstract. How a native gecko manages to coexist with an alien rodent in the Mediterranean since thousands of years? What kind of eco-ethological adaptations or evolutionary adjustments enables this gecko to persist? The present study explores the interaction between the endemic European Leaf-toed gecko (Euleptes europaea) and the alien Black rat (Rattus rattus). In the last 30 years, we compared 26 populations inhabiting “rat” and “rat-free” islands and islets in Tunisia, Sardinia, Corsica and Southern France. Geckos’ populations can persist despite the occurrence of rats. In the presence of rats: 1) geckos’ average body size tends to decrease towards medium-sized individuals; 2) geckos shift their spatial behaviour avoiding to forage “in the open”; 3) geckos’ body condition is not affected by the presence of rats. Moreover, shortly after rats’ eradication, geckos’ population structure seems to change and larger sized geckos prevail while the spatial behaviour is much more conservative. The mechanisms driving the interactions between the two spe- cies still need to be explained. Rats could represent a stressor for geckos, compete for space, be pest vectors and even predators. Coexistence of natives and aliens requires adaptive plasticity and evolutionary adjustments. In contexts where the risk of reinvasion is high, eradication programs need to be carefully evaluated, since the arrival of “new rats” on an island could have much more damaging effects on the insular biota than those caused by the eradicated population.
Keywords. Behavioural shift, disturbance, ecological plasticity, evolutionary processes, predation, rat eradication, Euleptes europea, Rattus rattus.
In memory of Michel Pascal (1947-2013) et al., 2013). In the last centuries, with a speeding up in the last decades human-induced dispersal of species has implied: a) huge acceleration of the introductions’ rate; INTRODUCTION b) great number of species involved; c) remote origins of some propagules; d) possible genetic modifications The long term dynamics of insular biota reflect the originated from captive breeding and farming activities. interplay between recurrent immigration and extinction Insularity-associated features, such as high sensitivity events (Brown and Lomolino, 2000). Mainly because of to climate and sea level changes, high rate of speciation, human expansion on global scale, the natural phenom- immigration and extinction processes, naiveté of native enon of sea dispersal has been greatly amplified favouring species, make insular biota particularly fragile (Moser et introductions and even biological invasions (Simberloff al., 2018).
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 90 Michel-Jean Delaugerre et alii
Over the last decades, conservation biology has cies, rats can dramatically impact on island biodiversity emphasized the detrimental effects of introduced species (Recher and Clark, 1974; Atkinson, 1985; Courchamp et (mostly mammals) on insular ecosystems, but little atten- al., 2003; Bennett et al., 2005) and Black rats in particu- tion has been paid to the understanding of the eco-etho- lar, have caused rapid extinctions on islands, as reported logical adaptations that enable a native species to persist for New Zealand and the Hawaiian Archipelago (Towns et despite the presence of aliens (Martin et al., 2000; Hoare al., 2006; Drake and Hunt, 2009). Alien rats can predate et al., 2007; Ruffino et al., 2009). In systems with novel and compete with native species and modify islands’ food combination of species (native and introduced ones) that chains (Traveset and Richardson, 2006). The worldwide do not show co-adaptation, rapid evolutionary changes success of the Black rat as colonizer is due to its ability to may occur (Berthon, 2015; Mooney and Cleland, 2001; exploit a large range of habitats and resources (Jones et al., Strauss et al., 2006; Sax et al., 2007; Stuart et al., 2014). 2008), as well as to shift to seasonal resources (Caut et al., On islands, in particular, morphological, ecological and 2008). This plasticity is crucial to survive in poor insular behavioural shift are more easily detectable than in more ecosystems, characterized by strong resource variation. complex systems (Simberloff, 1974). On small islets E. europaea and R. rattus are often Within the Mediterranean, reptiles, notably lizards, the only sedentary vertebrates. They are both nocturnal, are good tools to investigate insular evolution. These her- good climbers on rocky substrates, and they both forage petological communities, which are currently found on on rocky outcrops and on low vegetation. continental islands and on many land-bridge islets, origin Rat eradication programs are often focused on one from the neighbouring continents or large islands before or few emblematic species (Capizzi et al., 2010), without their isolation, dating back to the sea level raise of the a comprehensive understanding of insular assemblages. last interglacials. Trans-marine dispersal is less frequent Investigating the possible interactions between the Euro- in the Mediterranean arid islands (Foufopoulos and Ives, pean Leaf-toed gecko and the Black rat the present work 1999). could provide useful data for long-term insular conserva- Most of the endemic herpetofauna still remains on tion plans. some big islands (e.g., Corsica, Sardinia) and relative sat- In particular, we explored the effects of the Black rat ellite islands, while on other major Mediterranean islands on the Leaf-toad gecko population structure (analysed on the endemic herpetofauna has been completely replaced the basis of the relative proportion of size classes), body by human-introduced species, and endemic species per- condition and habitat use, comparing populations liv- sist only on some of their satellite islands (Corti et al., ing on islands colonized by rats and islands free of rats. 1999; Silva-Rocha et al., 2018, 2019). Because we expect rats to interfere with geckos’ habitat In the present work we investigated the interaction use, feeding habits and thermoregulation, geckos living between the native European Leaf-toed gecko Euleptes on islands on which also rats live should show: 1) pop- europaea (Gené, 1839) and the alien Black rat, Rattus rat- ulation structure with smaller-sized individuals result- tus (Linnaeus, 1758) on some small islands of the Western ing from a shorter lifespan; 2) poor body conditions; 3) Mediterranean. The European Leaf-toed gecko, a West- modification of spatial behaviour by minimizing or even ern Mediterranean endemic species listed in Appendix II avoiding foraging in open spaces. of the European Habitat Directive, is a very old inhabit- ant of this region, presumably since several million years ago (Müller, 2001). This species underwent a process of MATERIAL AND METHODS historic extinction or steep demographic decline along the North-Western (Provence, France) and Southern (Tuni- Ecological or evolutionary responses of native species to sia) edges of its range (Delaugerre et al., 2011). The Black aliens cannot be easily distinguished from the pre-existing eco- rat colonized the Western Mediterranean approximately logical differences (Strauss et al., 2006). However, even if each micro insular population differs from the others (e.g., due to 2000-2400 years ago (Thibault et al., 1987; Vigne and Val- age of isolation, presence of competitors and/or predators, etc.), ladas, 1996; Ruffino and Vidal, 2010). Thus, the coexist- any detectable trend can provide strong correlative evidences. ence between the two species might last since hundreds of geckos’ generations and thousands of rats’ generations, since Euleptes lives longer than rats (Salvidio et al., 2010 Study species and ref. therein). According to Ruffino et al. (2009), in the Western Mediterranean, 74% of the islands between 1 Euleptes europaea is a small gecko (average snout-vent and 5 ha (and 99% of islands larger than 30 ha) are colo- length, hereafter SVL, 38 mm; average adult weight 1.2 g; hatch- nized by Black rats. Among the human introduced spe- ling weight 0.25 g) endemic to the Western Mediterranean, mostly found on islands in rocky habitats. It is strictly noctur- Native gecko and Black rat on Mediterranean islands 91 nal, spending daytime in narrow rocky crevices (opening 2-4 = 47) consisted of 1 up to 6 nights per island (Table 1). Active mm wide) where dorsal and ventral parts of the body are in geckos were searched using battery-powered lamps starting one contact with the rock. These crevices are also used for egg lay- or two hours after dusk and lasted until dawn. Total sampling ing (Salvidio et al., 2010). Achieving its sexual maturity at the effort achieved more than 380 hours (Table 1). Geckos behav- age of 3 years (Salvidio and Delaugerre, 2003), it might live 6 iour was classified as follow: “in the open” when found on bare to 8 years in the wild and its maximum longevity in captivity rocks, on the ground or on plants; “under cover” when found attains 21 years (F. Molle in Mertens, 1970). hidden by the vegetation at the base of rocks (Fig. 2). Distance It feeds on flying and ground dwelling invertebrates. On from the ground (height of the first sight) was also measured. islets, densities are higher than on the mainland and popula- Geckos were carefully caught by hand and temporarily stored tion size ranges from several hundreds to only few dozen adults. in bags; sex and adulthood was determined according to This gecko is able to live on tiny islets (hundreds of square Delaugerre and Dubois (1985). Snout-vent length (SVL) of 1795 meters) characterized by the presence of few vascular plants, individuals was measured to the nearest 0.01 mm using a digi- where it represents the only sedentary vertebrate (Delaugerre tal calliper, weight (W) of 424 adults was recorded to the near- and Cheylan, 1992). Body size greatly varies on islets with est 0.01g using a digital scale. Geckos were released in the area trends towards gigantism, dwarfism and variation of the sexual of original sighting. Body condition index (BCI) was calculated size difference (Delaugerre and Cheylan, 1992; Salvidio et al., as from Bonnet and Naulleau (1994). Spatial behaviour of 1012 2010). geckos (i.e., “in the open” or “under cover”), recorded for 24 Rattus rattus (average weight 170 g), is a good climber, sampling sessions carried out on 17 islets, was assessed for one both on rocks and trees, but not a long lasting swimmer and hour “catch per-unit-effort” (CPUE). therefore considered with poor marine dispersal capacities (≤ 500 m) (Cheylan, 1988; Ruffino et al., 2009). In the Mediterra- nean, rats feed mainly on plants, avoiding or rarely eating halo- Statistical analyses philous and nitrophilous ones (Cheylan, 1988; Cassaing et al., 2007). Black rats and Euleptes europaea are both nocturnal and Population structure. In order to assess whether the popu- share the same habitat. lations structure of E. europaea was affected by the presence of rats, we compared gecko’s size between islands with and with- out rats. Since Leaf-toed geckos’ size varies among islands inde- Study sites and data collection pendently of rats, we firstly normalized body sizes by dividing each measure by the greatest observed value in order to have We focused on islands characterized by simple ecosystems, all values in the 0-1 range (Legendre and Legendre, 2012). Each where strong interactions between the target species are more individual was classified depending on its inclusion in the quar- likely to occur, namely small islands characterised by seasonal tile of the distribution of the normalized body size as “small” shortage of food availability for rats, and higher probability of (1st and 2nd quartiles), “medium” (3rd quartile), and “large” (4th encounters between the two species. We selected 26 islands and quartile). Geckos’ population structure was assessed by com- islets (Table 1) throughout most of the Leaf-toed gecko range puting the proportion of individuals per size class. Different (Fig. 1). Islands were classified as “rat” vs “rat-free”, depending population structures were computed for spring and autumn. on the presence or absence of rats; islands where rats were erad- Sample size included 1871 individuals from 24 islets (Mean val- icated were considered as “rat-free”. The presence of rats was ues ± SE: 78 ± 16); islands with less than 10 individuals meas- detected thanks to at least one of the following evidences: direct ured (San Bainzu, Piana and Porro) were excluded. Data were sightings, fresh and/or old faeces, remains of chewed olive seeds analysed using permutational multivariate analysis of variance, or plants, rats’ nests, rats’ urine scent. Past presence/absence of PERMANOVA (Anderson, 2001; McArdle and Anderson, 2001) rats on islands was investigated through literature and observa- on the basis of Euclidean distances among islands’ population tions made by locals. Nine islands (the largest ones) were inhab- structures. The PERMANOVA allows the multivariate informa- ited by rats probably since their early colonization (Abdelkrim tion to be partitioned according to the full experimental design et al, 2009), whereas 13 islands (the smallest and most remote without any a priori assumption regarding the distributions of ones, without edible plants) were presumably never inhabited the original variables. The predictors were rat occurrence (yes/ by rats. On Lavezzu Island (Corsica) we collected data before no), season (spring/autumn), number of reptile species, number and after rat eradication (performed in 2000): the island was of sampling sessions (to account for repeated measures within considered as “rat” before the eradication and as “rat-free” after an island), and island size. P-values were obtained by permuta- eradication (Table 1). Five islands with “transient” rat popula- tion, and the number of permutations was set to 9999. tions (i.e., small islets close to a colonization source but lacking Body condition. The effect of rat occurrence on the body resources to support a permanent rat population) were treated condition of geckos was assessed using a linear mixed model in as “rat”, because rat frequency on these islands is unknown. which the BCI was the dependent variable; rat occurrence, sex, and season were the fixed effects, whereas the island was the random factor accounting for repeated measuring. Sampling methods Spatial behaviour. Linear mixed models were also used to investigate the effect of rat occurrence on the habitat use by Observations were carried out in spring, summer and geckos. In a first analysis we checked if Leaf-toed geckos avoided autumn from July 1983 until August 2016. Sampling sessions (n 92 Michel-Jean Delaugerre et alii
Table 1. Map identification numbers (see Figure 1) of the study islands, geographical region, surface, presence (“rat”) or absence (“rat-free”) of the Black rat, and number of reptile species. For each island we also indicate sampling dates (and number of sampling nights), sampling effort (in minutes), number of geckos measured (n Biometry), number of geckos for which activity data were recorded (n Activity).
Year (month): Sampling n n n° Islet/island Region Surface (ha) Rat status N sp reptiles n nights effort (min) Biometry Activity 1 Gallo N_Tunisia 8.9 rat 2 2008 (05):3 1320 49 / rat 2010 (07):1 270 / 30 2 Toro Sardinia 13.22 rat free 3 2015 (06):1 195 34 45 3 Carpa Sardinia 0.4 rat 2 2011 (09):1 430 50 67 4 Porco Sardinia 4.8 rat 4 2012 (05):1 285 44 55 5 Spargiotto Sardinia 11.13 rat free 3 2014 (05):1 360 66 79 6* Lavezzu Corsica 62.73 rat 3 1986 (08):1 390 50 / rat free 2010 (06):3 550 40 41 rat free 2011 (06):3 427 67 41 rat free 2012 (06):2 720 66 / 7 Porraggia Grande Corsica 1.25 rat free 2 1985 (08):1 600 50 / 8 Porraggia piccola Corsica 0.61 rat free 2 1986 (08):1 410 59 / 9 Sperduto grande Corsica 0.92 rat free 1 1984 (10):2 785 40 / rat free 1986 (08):1 330 81 / rat free 2011 (06):1 435 43 43 10 Toro grande Corsica 1.62 rat free 2 1986 (08):1 360 59 / rat free 2005 (04):1 200 50 / rat free 2012 (07):1 140 / 52 rat free 2014 (07):1 250 48 59 1st islet NE of 11 Corsica 0.11 rat free 1 1986 (08):1 240 15 / Toro piccolo 12* Vacca Corsica 0.65 rat free 2 1985 (08):3 1395 95 / rat free 2012 (07):1 165 8 15 13 Roscana Corsica 0.2 rat free 1 1986 (08):1 720 94 / rat free 2008 (10):3 2258 125 / rat free 2012 (09):2 824 122 / 14 Locca Corsica 0.79 rat 2 2010 (04):1 90 / 10 15 Mezzumare Corsica 35.66 rat 3 2012 (08):1 213 15 18 16 A Botte Corsica 0.52 rat free 1 2010 (09):3 630 35 43 rat free 2011 (06):6 990 / 62 17 Gargalu Corsica 20.06 rat 4 1985 (04):2 1060 50 / rat 1990 (07):2 505 19 / 18 Palazzu Corsica 0.47 transient 1 1986 (08):1 225 26 / 19 Palazzinu Corsica 0.1 transient 1 1985 (07):2 490 34 / 20 Porri Corsica 0.27 rat free 1 1983 (07):2 590 32 / rat free 1986 (07):1 335 69 / 21 Brocciu Corsica 1.11 transient 1 2012 (06):1 132 30 60 22 Giraglia Corsica 10.35 rat free 4 2000 (09):1 360 38 / rat free 2012 (08):2 406 33 33 rat free 2012 (10):1 230 / 50 rat free 2014 (07):1 230 / 19 rat free 2014 (10):1 190 36 38 rat free 2015 (08):1 220 / 32 23 Saint Ferreol Provence 0.45 rat 2 2016 (05):1 700 23 40 24 La Tradelière Provence 1.05 rat 2 2016 (05):1 460 23 37 25 Rascas Provence 0.76 transient 2 1985 (09):1 290 50 / 26* Gabinière Provence 3.42 transient 2 2003 (10):1 345 69 / transient 2016 (05):1 187 37 43
* On Lavezzu Island (n. 6) rats were eradicated in 2000; on the Toro islets (n. 10 and 11) in 1992, after 2-4 years of presence on the islets; on Vacca (n. 12) in January 2011, after having been detected in July 2010; on Gabinière (n. 26) rats were present in 1937, eradicated in 1966, rats re-invaded the island in 2010 and were eradicated again in 2014. The islets San Bainzu, Piana and Porro, all inhabited by rats, were escluded from the analysis because less than 10 geckoes were measured. Native gecko and Black rat on Mediterranean islands 93
included 16 islands for which we collected more than 10 individ- uals (n = 1001, Mean values ± SE: 62 ± 10) for computing CPUE indexes for both “in the open” and “under cover” microhabitats. In a second analysis we checked if the height above the ground of the first sighting differed in islands with or without rats, using rat occurrence, season, number of reptile species, island size, and the rat occurrence × season as fixed predictors. As in the previ- ous analysis, the islands entered the model as random effect, and geckos observed “under cover” were excluded. The sample for this last analysis included 469 geckos from 14 islands. Analyses were performed using the package lme4 (Bates et al., 2014) in R ver. 3.2.4 (R Core Team, 2018), and otherwise stated, data reported are means ± standard error.
RESULTS
Population structure
The PERMANOVA showed that the population struc- ture significantly varied in response to rat occurrence, Fig. 1. Geographic range of Euleptes europaea and studied popula- season, number of reptile species living on the island, but tions. Numbers refers to Table 1. was invariant in respect to the sampling effort (Table 2). In particular, a) the structure of geckos’ populations liv- ing on the islands with rats showed a larger proportion of medium-sized individuals and a lower proportion of large-sized geckos compared to populations occurring on “rat-free” islands (Fig. 3a); b) irrespective of rat occur- rence, smaller geckos were frequent in spring, while larger ones prevailed in autumn, and the relative abundance of medium-sized individuals did not vary between seasons (Fig. 3b); c) the greater was the number of reptile spe- cies on the island, the smaller was the frequency of larger geckos and the higher the number of medium-sized indi- viduals (Fig. 3c). Moreover, large geckos were more fre- quent on islets rather than on islands (Fig. 3d).
Body condition Body condition of the Leaf-toed geckos did not dif- Fig. 2. Vegetation (mostly Lotus cytisoides) covering the base of the fer between “rat” and “rat-free” islands (0.0292 ± 0.0005 granite boulders inhabited by Euleptes europaea on Spargiotto Islet (Maddalena Archipelago). May 2012. Table 2. Results of permutational ANOVA for the variability of the population structure of the Leaf-toed geckos in response to rat occurrence, season and number of reptile species after having been “in the open” microhabitats on islands with rats: the catch per controlled for island size and number of sampling years. P-values unit effort (CPUE) was the dependent variable, while rat occur- are computed with 9999 permutations. rence, habitat type (“in the open” vs “under cover”), island size and the number of reptile species were the fixed predictors. Variable df F R2 P We also added the rat occurrence × habitat type interaction to Rats 1,24 3.916 0.08 0.0374 account for differential habitat use between “rat” and “rat-free” Islet size 1,24 3.775 0.08 0.0374 islands. We could not include in the model the season as all but three islands were sampled in spring, and the island entered the Season 1,24 10.69 0.23 0.0010 model as random effect to control for double measuring within N. reptiles 1,24 3.975 0.08 0.0336 island (i.e., CPUE in open and close microhabitat). Sample size N. replicates 1,24 0.0033 <0.01 0.9958 94 Michel-Jean Delaugerre et alii
Table 3. Results of the linear mixed model for the variability of the body condition index of male and female Leaf-toed geckos in response to rat occurrence. Only fixed effects are reported.
Predictor df F P Rat occurrence 1,6 0.0053 0.94 Season 1,6 0.608 0.46 Sex 1,5 2.545 0.11
Table 4. Results of the linear mixed model for the variability of the activity of Leaf-toed geckos in response to rat occurrence (df have been calculated using the Satterthwaite approximation). Only fixed effects are reported.
Predictor df F P Open vs close habitats Rat occurrence 1,12.7 0.781 0.39 Habitat type 1,24.9 1.344 0.26 N. reptile species 1,8.5 0.629 0.45 Island size 1,8.3 0.802 0.39 Fig. 3. Variation of the population structure of the European Leaf- toed gecko on 22 Mediterranean islands in response to (a) rat Rat occurrence × Habitat type 1,24.9 11.99 0.0019 occurrence (present in yellow, absent in green), (b) season (Spring Height above ground in yellow, Autumn in green), (c) number of reptile species living Rat occurrence 1,14.9 1.068 0.32 on the islands (1 sp. in red, 2 spp. in yellow, 3 spp. in green, 4 spp. Season 1,22.3 0.659 0.42 in blue), and (d) island size (islet: < 10,000 m² in yellow, island: > 10,000 m² in green). Bars = 95% confidence interval. N. reptile species 1,2 0.142 0.74 Island size 1,4.4 2.168 0.21 Rat occurrence × Season 1,19.2 2.446 0.13 and 0.0285 ± 0.0002 respectively, F1,5 = 0.0053, P = 0.94), either between males and females (0.0282 ± 0.0003 and 0.0292 ± 0.0004 respectively, F1,413 = 2.545, P = 0.11), or nificant, suggesting that CPUE was not affected by island season (autumn: 0.0269 ± 0.0004; spring: 0.0295 ± 0.0003, features other than rat occurrence and habitat. F1,5 = 0.608, P = 0.46) (Table 3). By contrast, the random By contrast, the height from the ground where the effect (island) was highly significant (L-ratio χ² = 24.0, d.f. geckoes were observed “in the open” was not affected = 1, P < 0.001), and the effect (σ = 0.0021) accounted for by the rat occurrence (“rat-free” islands: 47.6 ± 2.6; “rat” 32% of the total variance, suggesting that body condition islands: 32.6 ± 6.5), either by season (autumn: 51.7 ± 5.0; is highly dependent on the island features. spring: 44.3 ± 2.7) or by their interaction (Table 4). Simi- larly, the effects of both island size and the number of rep- tile species were also negligible (see Table 4 for statistics). Spatial behaviour We report some additional observation on the spa- tial behaviour. Before rat eradication, on Lavezzu Island Leaf-toed geckos were more active on “rat-free” in 1982, on 112 sightings ≈85% of geckos were active islands than on “rat” islands (values of CPUE being 7.85 under plant cover; in 1986 (n = 50) the proportion was ± 2.35 and 6.33 ± 2.30, respectively), but this difference ≈80-85% (MD pers. obs.). Ten years after rat eradication, was significantly depending on the habitat (rat occurrence the percentage of geckos active under plant cover was × habitat type interaction: F = 11.99, P = 0.0019, Table 1,24 lower but still high: 67% in 2010 (n = 41), 68% in 2011 4). Geckos on “rat-free” islands were active “in the open” (n = 41). On the remote Vacca Islet, rats were detected rather than “under cover” (P = 0.057), while the opposite for the first time in 2010 and eradicated by the Bouches occurred on “rat” islands (P = 0.0091, Fig. 4), suggesting de Bonifacio Natural Reserve in less than one year (O. that the Leaf-toed geckos actually avoid insecure micro- Bonnenfant pers. com.). We do not know the activity pat- habitats in presence of rats. All other predictors were not tern before eradication, but in July 2012, 87% of geckos significant (see Table 4 for details). The random effect of were observed in the open. Analogously, few miles away, islands (L-ratio χ² = 2.92, d.f. = 1, P = 0.09) was not sig- Native gecko and Black rat on Mediterranean islands 95
Gasc et al., 2010); ascertained (Harper and Bunbury, 2015; Thibault et al., 2016; Clapperton et al. 2019) or observed for other rat species (Towns et al., 2006). Abun- dance of lizards (and Tuataras) has been related to the presence of rats but the mechanisms still deserve to be explained and likely could involve competitive processes less obvious than predation (Towns et al., 2006). Other Vertebrates or invertebrates living on the studied islands could interact with the Leaf-toed gecko as possible competitors and/or predators such as: the ant Crematogaster scutellaris, competing for rock crev- ices and predating hatching geckos (Delaugerre, 1981); introduced cows, grazing on vegetation used by geckos foraging “under cover”, e.g., on Lavezzu (Delaugerre and Brunstein, 1987); the bird Monticola solitarius and the ant Fig. 4. Variation of the foraging mode of the European Leaf-toed Tapinoma erraticum predating E. europaea respectively geckos, “in the open” (open) and “under cover” (close) in response on Lavezzu (Delaugerre and Cheylan, 1992) and Porrag- to the presence of rats. The relative activity of geckoes was obtained from sightings of one hour of “catch per-unit-effort” (CPUE) on 15 gia Grande islands (Delaugerre and Brunstein, 1987) and islets. Bars = 95% confidence interval. perhaps the Western Whip snake Hierophis viridiflavus, that on Giraglia Island has been observed to have noctur- nal habits (Delaugerre, 2013). However 73% of the stud- rats invaded the Toro Islands where they have stayed for ied islands, host just one or two reptile species (Table 1), about 4 years, until 1992, when eradication occurred the second species being typically a Podarcis (diurnal) liz- (Thibault, 1992). In April 2005 (n = 50) ≈90%, in July ard, that is unlikely a predator of the Leaf-toed gecko. On 2012 (n = 52) 96% and in July 2014 (n = 59) 93% geckos the remaining 27% islands, the presence of three or four were observed in the open. (just on three islands) reptile species is equally distribut- ed between rat and rat-free islands. Alternatively, rats may affect gecko’s population DISCUSSION structure by interfering with their growth process. Sur- vival of geckos can be reduced by increased physiological Two out of the three predictions on how the Black stress, which ultimately may cause physiological shifts in rat may influence the activity and demography ofEuleptes life history strategy and demographic fitness components europaea were confirmed by our results: a) in the pres- (Rödl et al., 2007; Trompeter and Langkilde, 2011; Naray- ence of rats smaller sized geckos prevail, and b) geckos an et al., 2013). Rats could act as stressors for geckos just are less active “in the open”; c) by contrast, no significant roaming around. Moreover, geckos could be subjected effect on body condition was found. to infections carried by black rats (Prenter et al., 2004), Transition to populations characterized by smaller often infested by the adult nematode Mastophorus muris sized individuals could be caused by selective preda- (Cassaing et al., 2005). According to Lafferty et al. (2010), tion on larger geckos. On Ohınau Islet in New Zealand, native geckos of the central Pacific islands might serve juveniles of the giant gecko Hoplodactylus duvaucelii are as paratenic hosts of M. muris; this stage being the most more vulnerable to predation by Rattus exulans (Hoare et incline in affecting host fitness (Kuris, 2003). al., 2007), likely because adult geckos are similar in size Rats might also interact with geckos in a more to the Pacific rat. Conversely, a 170 g Black rat could eas- indirect way, for instance inducing vegetation changes. ily predate European Leaf-toed geckos of any size (from Lotus cytisoides, for example, is a plant that provides 0.3 to 2 g). If this would be the case, a random predation high quality shelters for geckos and for a lot of inverte- of any size class could ultimately also result in a shortage brates (geckos’ preys) and it is consumed by rats (Cas- of the larger sized geckos. In the Mediterranean, despite saing et al., 2005; Ruffino et al., 2011). Behavioural shift some detailed studies on this topic (Cassaing et al., 2005; in microhabitat use might be costly for geckos, since Pérez-Mellado et al., 2008; Petralia et al., 2010; Ruffino et they would be forced to forage under plant cover with al., 2011), direct predation of Rattus rattus on Sauria was consequently limited access to bare rock surfaces more never observed, although predation on lizards has been favourable for nocturnal thermoregulation (thigmother- supposed to occur in other regions (Caut et al., 2008; my). 96 Michel-Jean Delaugerre et alii
Information gathered on Lavezzu before and after rat Savidge et al., 2012). Once settled on an island, a rat eradication seems to confirm an influence of the presence population may repel new invading rats through aggres- of rats on geckos’ population structure. Before rat remov- sive interactions, as observed by Granjon and Cheylan al, as observed for other islands with rats, geckos’ size (1989) and Abdelkrim et al. (2009) who found that the class structure was skewed towards medium-sized indi- Lavezzu population was founded by a single coloniza- viduals. A decade after eradication (three geckos genera- tion event without further gene flow. Hence, eradication tions) we observed an upward trend towards larger sized might entice new invaders to settle in a competition-free individuals, even if the sample size of some size classes is ecosystem. too small to perform a statistical test. Moreover, island “old-resident” rat populations are Rats seem to change geckos’ population structure adapted to exploit seasonal resources adjusting home without affecting body condition of individuals. Indeed, range, demography and intraspecific interactions (Chey- geckos’ body condition was not influenced by the pres- lan, 1988; Clark, 1980; Cassaing et al., 2007; Ruffino et al., ence of rats, sex or season. However, we found a strong 2011; Pisanu et al., 2011). New invading rats (e.g., main- significant random effect of islands, suggesting that body land rats), not familiar with the island conditions, might condition in both males and females strongly depends on severely affect the local biota. Observations carried out some island features rather than on the presence of rats. on the Corsican Toro islands 2-4 years after rat invasion, Activity, assessed by the catch per unit effort, did not besides the well-known detrimental effects on the Cory’s vary just according to rat occurrence, but also depending shearwater and on the Pallid swift, revealed that rats on the microhabitat. In the presence of rats, geckos avoid- preyed almost to extinction Silene vellutina (Caryophyl- ed to forage “in the open” and were more active “under laceae) (Thibault, 1992), a rare endemic plant that else- cover”, probably to avoid disturbance and/or predation risk where was not (or only marginally) consumed by “old- induced by the ground-dwelling rodents (Whitaker, 1973; resident” rats. Hoare et al., 2007). On Lavezzu Island, the “under cover” Since several eradication actions have ‘failed’, and was the most common mode both before and after rat reinvasions occurred (Cheylan and Granjon, 1987; How- eradication, even if the percentages of geckos active “under ald et al., 2007; Russell et al., 2010; Savidge et al., 2012; cover” decreased ten years after eradication. This pattern Sposimo et al., 2012; Ragionieri et al., 2013), the urge to (even if related to one island) may suggest a certain persis- learn from failures led to the proposal to adopt the meta- tence of an avoidance behaviour likely resulted from a pro- populational (Russell et al., 2008) or the eradication units longed coexistence with rodents. Though in geckos as well approach (Robertson and Gemmell, 2004) by considering as in most reptiles, young individuals cannot learn through first islands’ assemblages and relative reinvasion sources parental care or imitation, the acquisition of a novel behav- in order to properly guide eradication programs. Native iour might take long time. Unlike how observed for the island species must cope with an increasing number of New Zealand Hoplodactylus duvaucelii who recovered its biological invasions; understanding the mechanisms of arboreal habitat six months after rat removal (Hoare et invasions, as well as the interaction between native spe- al., 2007), Lavezzu Euleptes gecko did not show a similar cies and long coexisting aliens, may be crucial to adopt behavioural plasticity. Dealing with antipredator behav- proper conservation actions. iours on islands Blumstein (2002) reported that «experi- ence dependant behaviours change rapidly following iso- lation» (and the loss of predator), «whereas more hard ACKNOWLEDGMENTS wired behaviours may persist for many generations.» Does Euleptes geckos have acquired an avoidance behaviour We would like to thank: Joanne Monks (born Hoare), over the last 2000 years while coexisting with Black rats? who kindly granted us to borrow part of the title and Or, does this behaviour date back to the Pleistocene when structure of the paper (Hoare et al., 2007); Initiative geckos coexisted with mammals nowadays extinct (e.g., the Petites Iles de Méditerranée (PIM), APAL (Tunisia), CEN Tyrrhenian field rat, Rhagamys orthodon, M. Masseti and Corse, Réserves naturelles des Bouches de Bonifacio, G. Cheylan (pers. comm.)? However, geckos’ behavioural de Scandola et des Iles de la Pointe du Cap Corse, Parc changes due to the presence of rats seem to be very vari- naturel régional de la Corse, Parco Nazionale dell’ Arci- able (Hoare et al., 2007; Krebs et al., 2015). pelago de La Maddalena, Parc National de Port-Cros; the As suggested by our results, geckos may benefit colleagues and friends who helped during field work: A. from rat eradication in most contexts but a lot of cau- Abiadh, Ch.-H. Bianconi, V. Bosc, D. Brunstein, M.H. tion should be paid when undertaking rat eradications on Casalonga, G. Deso, Y. Donno, A. Gaio, O. Gerriet, F. islands that likely can be reinvaded (Harris et al., 2011; Grita, N. Nègre-Santucci, O. Patrimonio, J. Renet. Native gecko and Black rat on Mediterranean islands 97
Our study was improved by valuable discussions with taire du rat noir Rattus rattus dans les îles d’Hyères Gilles Cheylan, Marc Cheylan, Marco Masseti, Michel analysé par la biochimie isotopique et les contenus Pascal and Jean-Claude Thibault. stomacaux. Sci. Rep. Port-Cros Natl. Park, Fr. 21: Permits to handle protected species were issued 85-115. by the Dreal (Corse and PACA) and by the Ministero Cassaing, J., Derré, C., Moussa, I., Cheylan, G. (2007): dell’Ambiente e della Tutela del Territorio e del Mare (U. Diet variability of Mediterranean insular populations prot. DPN 0010637B 16/05/2009, 0017564 12/08/2010, of Rattus rattus studied by stable isotope analysis. Iso- 0044068 04/12/2012 and 0001805 04/02/2015) and the Par- topes in Environmental and Health Studies 43: 197- co Nazionale dell’Arcipelago di La Maddalena. The access to 213. Giraglia Island was authorized by the Préfet de Haute-Corse. Caut, S., Angulo, E., Courchamp, F. (2008): Dietary shift of an invasive predator: rats, seabirds and sea turtles. J. Appl. Ecol. 45: 428-437. REFERENCES Cheylan, G. (1988). Les adaptations écologiques de Rat- tus rattus à la survie dans les îlots méditerranéens Abdelkrim, J., Pascal, M., Samadi, S. (2009): Genetic (Provence et Corse). Bull. Ecol. 19: 417-426. structure and functioning of alien ship rat popula- Cheylan, G., Granjon, L. (1987). Ecologie du rat noir à tions from a Corsican micro-insular complex. Biol. Lavezzi (Corse du Sud): abondances, déplacements Invasions 11: 473-482. et reproduction. Travaux Scientifiques―Parc Naturel Anderson, M.J. (2001): A new method for non‐paramet- Régional et Réserves Naturelles de Corse 12: 71-91. ric multivariate analysis of variance. Austral Ecol. 26: Clapperton, K., Maddigan, F., Chinn, W., Murpy, E. 32-46. (2019): Diet, Population Structure and Breeding of Atkinson, I.A. (1985): The spread of commensal species Rattus rattus L. in South Island Beech Forest. New of Rattus to oceanic islands and their effects on island Zeal. J. Ecol. 43: 1-8. avifaunas. Conservation of Island Birds 3: 35-81. Clark, D.B. (1980): Population ecology of Rattus rattus Bates, D., Mächler, M., Bolker, B., Walker, S. (2014): Fit- across a desert-montane forest gradient in the Galápa- ting linear mixed-effects models using lme4. ArX- gos Islands. Ecology 61: 1422-1433. iv:1406.5823. Corti, C., Böhme, W., Delfino, M., Masseti, M. (1999): Bennett, P.M., Owens, I.P.F., Nussey, D., Garnett, S.T., Man and lacertids on the Mediterranean islands: con- Crowley, G.M. (2005): Mechanisms of extinction in servation perspectives. Natura Croatica 8: 287-300. birds: phylogeny, ecology and threats. In: Phylogeny Courchamp, F., Chapuis, J.-L., Pascal, M. (2003): Mam- and Conservation (Conservation Biology), pp. 317- mal invaders on islands: impact, control and control 336. Purvis, A., Gittleman, J., Brooks, T., Eds, Cam- impact. Biol. Rev., 78: 347-383. bridge University Press, Cambridge. Delaugerre, M. (1981). Sur l’histoire naturelle de Phyllo- Berthon, K. (2015): How do native species respond to dactylus europaeus Gené, (Gekkonidae, Sauria, Rep- invaders? Mechanistic and trait-based perspectives. tiles): Port-Cros : étude d’une population naturelle. Biol. Invasions 17: 2199-2211. Trav. Sci. Parc. Nation. Port-Cros 6: 147-175. Blumstein, D.T. (2002): Moving to suburbia: ontogenetic Delaugerre, M., Brunstein, D. (1987): Observations sur la and evolutionary consequences of life on predator- flore et la faune de plusieurs îlots du sud de la Corse free islands. J. Biogeogr. 29: 685-692. (archipels des Lavezzi, des Cerbicale et côte sud-ori- Bonnet, X., Naulleau, G. (1994): Utilisation d’un indi- entale). Travaux Scientifiques―Parc Naturel Régional ce de condition corporelle (BCI) pour l’étude de la et Réserves Naturelles de Corse 12: 1-17. reproduction chez les serpents. Comptes Rendus de Delaugerre, M., Cheylan, M. (1992): Atlas de répartition l’Académie Des Sciences. Série 3, Sciences de La Vie des batraciens et reptiles de Corse. Ajaccio Parc naturel 317: 34-41. régional de Corse: Ecole pratique des hautes études. Brown, J.H., Lomolino, M.V. (2000): Concluding remarks: Delaugerre, M., Dubois, A. (1985): La variation historical perspective and the future of island bioge- géographique et la variabilité intrapopulationnelle ography theory. Global Ecol. Biogeogr. 9: 87-92. chez Phyllodactylus europaeus (Reptilia, Sauria, Gek- Capizzi, D., Baccetti, N., Sposimo, P. (2010): Prioritizing rat konidae). Bulletin Du Muséum National d’histoire eradication on islands by cost and effectiveness to pro- Naturelle. Section A, Zoologie, Biologie et Écologie tect nesting seabirds. Biol. Conserv. 143: 1716-1727. Animales, 7: 709-736. Cassaing, J., Derré, C., Moussa, I., Parghentanian, T., Delaugerre, M., Ouni, R., Nouira, S. (2011). Is the Euro- Bocherens, H., Cheylan, G. (2005): Le régime alimen- pean Leaf-toed gecko Euleptes europaea also an Afri- 98 Michel-Jean Delaugerre et alii
can? Its occurrence on the Western Mediterranean nucifera) Habitat at Palmyra Atoll. J. Parasitol. 96: landbrige islets and its extinction rate. Herpetol. 16-20. Notes 4: 127-137. Legendre, P., Legendre, L.F. (2012): Numerical ecology. Delaugerre, M.-J. (2013). Going out tonight? When insu- Elsevier B.V., Amsterdam. lar Hierophis viridiflavus breaks the Whip Snakes Martin, J.-L., Thibault, J.-C., Bretagnolle, V. (2000): Black Rules. Acta Herpetol. 8: 47-52. rats, island characteristics, and colonial nesting birds Drake, D.R., Hunt, T.L. (2009): Invasive rodents on in the Mediterranean: consequences of an ancient islands: integrating historical and contemporary ecol- introduction. Conserv. Biol. 14: 1452-1466. ogy. Biol. Inv. 11: 1483-1487. McArdle, B.H., Anderson, M.J. (2001): Fitting multivariate Foufopoulos, J., Ives, A.R. (1999). Reptile extinctions on models to community data: a comment on distance- land-bridge islands: life-history attributes and vulner- based redundancy analysis. Ecology 82: 290-297. ability to extinction. Am. Nat. 153: 1-25. Mertens, R. (1970): Ueber die Lebensdauer einiger Gasc, A., Duryea, M.C., Cox, R.M., Kern, A., Calsbeek, R. Amphibien und Reptilien in Gefangenschaft. Zool. (2010): Invasive predators deplete genetic diversity of Gart. 39: 13-209 island lizards. PloS One 5: e12061. Mooney, H.A., Cleland, E.E. (2001): The evolutionary Granjon, L., Cheylan, G. (1989): Le sort de rats noirs impact of invasive species. P. Natl. Acad. Sci. USA, 98: (Rattus rattus) introduits sur une île, relevé par radio- 5446-5451. tracking. Comptes Rendus de l’Académie Des Scienc- Moser, D., Lenzner, B., Weigelt, P., Dawson, W., Kreft, es. Série 3, Sciences de La Vie 309: 571-575. H., Pergl, J., Pyšek, P., van Kleunen, M., Winter, M., Harper, G.A., Bunbury, N. (2015): Invasive rats on tropi- Capinha, C., Cassey, P., Dullinger, S., Economo, E.P., cal islands: Their population biology and impacts on García-Díaz, P., Guénard, B., Hofhansl, F., Mang, T., native species. Global Ecol. Conserv. 3: 607-627. Seebens, H., Essl, F. (2018): Remoteness promotes Harris, D.B., Gregory, S.D., Bull, L. S., Courchamp, F. biological invasions on islands worldwide. P. Natl. (2011): Island prioritization for invasive rodent eradi- Acad. Sci. USA, 115: 9270-9275. cations with an emphasis on reinvasion risk. Biol. Inv. Müller, J. (2001): A new fossil species of Euleptes from 14: 1251-1263. the early Miocene of Montaigu, France (Reptilia, Gek- Hoare, J.M., Pledger, S., Nelson, N.J., Daugherty, C.H. konidae). Amphibia-Reptilia 22: 341-348. (2007): Avoiding aliens: Behavioural plasticity in Narayan, E.J., Cockrem, J.F., Hero, J.-M. (2013): Sight of habitat use enables large, nocturnal geckos to survive a Predator Induces a Corticosterone Stress Response Pacific rat invasions. Biol. Conserv. 136: 510-519. and Generates Fear in an Amphibian. PLoS One 8: Howald, G., Donlan, C.J., Galván, J.P., Russell, J.C., e73564. Parkes, J., Samaniego, A., Wang, Y., Veitch, D., Geno- Pérez-Mellado, V., Hernández-Estévez, J.Á., García-Díez, vesi, P., Pascal, M., Saunders, A., Tershy, B. (2007): T., Terrassa, B., Ramón, M.M., Castro, J., Picornell, A., Invasive Rodent Eradication on Islands. Conserv. Biol. Martín-Vallejo, J., Brown, R. (2008): Population densi- 21: 1258-1268. ty in Podarcis lilfordi (Squamata, Lacertidae), a lizard Jones, H.P., Tershy, B.R., Zavaleta, E. ., Croll, D.A., Keitt, species endemic to small islets in the Balearic Islands B.S., Finkelstein, M.E., Howald, G.R. (2008): Severity (Spain). Amphibia-Reptilia 29: 49-60. of the Effects of Invasive Rats on Seabirds: A Global Petralia, E., Messina, A., Petralia, A., Siracusa, A.M. Review: Effects of Rats on Seabirds. Conserv. Biol. 22: (2010): Stato della popolazione di Rattus rattus (Lin- 16-26. naeus, 1758)(Rodentia, Muridae) nella Riserva Natu- Krebs, É, Abba, A, Gillet, P, Eudeline, R, Gauthier, J, Le rale Integrale “Isola Lachea e Faraglioni dei Ciclopi” Quilliec, P, Lorvelec, O, Martinerie, G, Vidal, É, Buis- di Acicastello (Catania, Italia). Bolletino Accademia son, É. (2015): Réponses des populations de reptiles Gioenia Sci. Nat., 43: 12-22. à l’éradication du Rat noir (Rattus rattus) sur l’île de Pisanu, B., Caut, S., Gutjahr, S., Vernon, P., Chapuis, J.-L. Bagaud (Parc national de Port-Cros, Var, France). Rev. (2011): Introduced black rats Rattus rattus on Ile de la Ecol.-Terre Vie. 70: 99-109. Possession (Iles Crozet, Subantarctic): diet and troph- Kuris, A.M. (2003): Evolutionary ecology of trophically ic position in food webs. Polar Biol. 34: 169-180. transmitted parasites. J. Parasitol. 89: S96–S100. Prenter, J., MacNeil, C., Dick, J.T., Dunn, A.M. (2004): Lafferty, K.D., Hathaway, S.A., Wegmann, A.S., Ship- Roles of parasites in animal invasions. Trends Ecol. ley, F.S., Backlin, A.R., Helm, J., Fisher, R.N. (2010): Evol., 19: 385-390. Stomach Nematodes (Mastophorus Muris) in Rats R Core Team. (2018). R: A language and environment for (Rattus rattus) Are Associated with Coconut (Cocos statistical computing. https://www.r-project.org/ Native gecko and Black rat on Mediterranean islands 99
Ragionieri, L., Cutuli, G., Sposimo, P., Spano, G., Navone, evolutionary insights from species invasions. Trends A., Capizzi, D., Baccetti, N., Vannini, M., Fratini, S. Ecol. Evol. 22: 465-471. (2013): Establishing the eradication unit of Molara Silva-Rocha, I., Montes, E., Salvi, D., Sillero, N., Mateo, Island: a case of study from Sardinia, Italy. Biol. Inv. J.A., Ayllón, E., Pleguezuelos, J.M., Carretero, M.A. 15: 2731-2742. (2018): Herpetological History of the Balearic Islands: Recher, H.F., Clark, S.S. (1974): A biological survey of When Aliens Conquered These Islands and What to Lord Howe Island with recommendations for the con- Do Next. In: Histories of Bioinvasions in the Mediter- servation of the island’s wildlife. Biol. Conserv. 6: 263- ranean, pp. 105–131. Queiroz, A.I., Pooley, S., Eds., 273. Springer International Publishing, Cham, Switzerland. Robertson, B.C., Gemmell, N.J. (2004): Defining eradica- Silva-Rocha, I., Salvi, D., Carretero, M.A., Ficetola, G.F. tion units in pest control programmes. J. Appl. Ecol. (2019): Alien Reptiles on Mediterranean Islands: A 41: 1032-1041. Model for Invasion Biogeography. Divers. Distrib. 25: Rödl, T., Berger, S., Romero, L.M., Wikelski, M. (2007): 995-1005. Tameness and stress physiology in a predator-naive Simberloff, D., Martin, J.L., Genovesi, P., Maris, V., War- island species confronted with novel predation threat. dle, D.A., Aronson, J., Courchamp, F., Galil, B., P. Roy. Soc. B-Biol. Sci. 274: 577-582. García-Berthou, E., Pascal, M., Pyšek, P., Sousa, R., Ruffino, L., Bourgeois, K., Vidal, E., Duhem, C., Paracuel- Tabacchi, E., Vilà, M. (2013): Impacts of biological los, M., Escribano, F., Sposimo, P., Baccetti, N., Pascal, invasions: what’s what and the way forward. Trends M., Oro, D. (2009): Invasive rats and seabirds after Ecol. Evol. 28: 58-66. 2,000 years of an unwanted coexistence on Mediterra- Simberloff, D.S. (1974): Equilibrium theory of island bio- nean islands. Biol. Inv. 11: 1631-1651. geography and ecology. Annu. Rev. Ecol. Syst. 5: 161- Ruffino, L., Russell, J.C., Pisanu, B., Caut, S., Vidal, E. 182. (2011): Low individual-level dietary plasticity in an Sposimo, P., Spano, G., Navone, A., Fratini, S., Ragionieri, island-invasive generalist forager. Popul. Ecol. 53: L., Putzu, M., Capizzi, D., Baccetti, N. (2012): Rodent 535-548. eradication on Molara Island and surrounding islets Ruffino, L., Vidal, E. (2010): Early colonization of Medi- (NE Sardinia): from success to the riddle of reinva- terranean islands by Rattus rattus: a review of zooar- sion. Aliens 32: 33-38. cheological data. Biol. Inv. 12: 2389-2394. Strauss, S.Y., Lau, J.A., Carroll, S.P. (2006): Evolutionary Russell, J.C., Beaven, B.M., MacKay, J.W.B., Towns, D.R., responses of natives to introduced species: what do Clout, M.N. (2008): Testing island biosecurity systems introductions tell us about natural communities? Ecol. for invasive rats. Wildlife Res. 35: 215. Lett. 9: 357-374. Russell, J.C., Miller, S.D., Harper, G.A., MacInnes, H.E., Stuart, Y.E., Campbell, T.S., Hohenlohe, P.A., Reynolds, Wylie, M.J., Fewster, R.M. (2010): Survivors or rein- R.G., Revell, L.J., Losos, J.B. (2014): Rapid evolution vaders? Using genetic assignment to identify invasive of a native species following invasion by a congener. pests following eradication. Biol. Inv. 12: 1747-1757. Science 346: 463-466. Salvidio, S., Delaugerre, M. (2003): Population dynamics Thibault, J.C. (1992): Eradication of the Brown Rat from of the European leaf-toed gecko (Euleptes europaea) the Toro Islets (Corsica): remarks about an unwanted in NW Italy: implications for conservation. Herpetol. colonizer. Avocetta 16: 114-117. J. 13: 81-88. Thibault, J.C., Delaugerre, M., Cheylan, G., Guyot, I., Salvidio, S., Lanza, B., Delaugerre, M.J. (2010): Eulept- Miniconi, R. (1987): Les vertébrés terrestres non es europaea (Gené, 1839). In: Fauna d’Italia Vol. 45 domestiques des îles Lavezzi (sud de la Corse). Bulle- (Reptilia), pp. 258-270. Corti, C., Capula, M., Luiselli, tin de La Société Linnéenne de Lyon 56: 73-103. L., Razzetti, E., Sindaco, R., Eds, Edizioni Calderini de Thibault, M., Brescia, F., Jourdan, H., Vidal, E. (2016): Il Sole 24 ORE, Milano. Invasive rodents, an overlooked threat for skinks in a Savidge, J.A., Hopken, M.W., Witmer, G.W., Jojola, S.M., tropical island hotspot of biodiversity. New Zealand J. Pierce, J.J., Burke, P.W., Piaggio, A.J. (2012): Genetic Ecol. 41: 74-83. evaluation of an attempted Rattus rattus eradication Towns, D.R., Atkinson, I.A.E., Daugherty, C.H. (2006): on Congo Cay, U.S. Virgin Islands, identifies impor- Have the harmful effects of introduced rats on islands tance of eradication units. Biol. Inv. 14: 2343-2354. been exaggerated? Biol. Inv. 8: 863-891. Sax, D.F., Stachowicz, J.J., Brown, J.H., Bruno, J.F., Daw- Traveset, A., Richardson, D. (2006): Biological inva- son, M.N., Gaines, S.D., Grosberg, R.K., Hastings, A., sions as disruptors of plant reproductive mutualisms. Holt, R.D., Mayfield, M.M. (2007): Ecological and Trends Ecol. Evol. 21: 208-216. 100 Michel-Jean Delaugerre et alii
Trompeter, W.P., Langkilde, T. (2011): Invader danger: Lizards faced with novel predators exhibit an altered behavioral response to stress. Horm. Behav. 60: 152- 158. Vigne, J.-D., Valladas, H. (1996): Small mammal fossil assemblages as indicators of environmental change in northern Corsica during the last 2500 years. J. Archaeol. Sci. 23: 199-215. Whitaker, A.H. (1973): Lizard populations on islands with and without Polynesian rats, Rattus exulans (Peale). Proc. New Zeal. Ecol.Soc. 20: 121-130. Acta Herpetologica 14(2): 101-107, 2019 DOI: 10.13128/a_h-7747
PIT-Tags as a technique for marking fossorial reptiles: insights from a long-term field study of the amphisbaenian Trogonophis wiegmanni
Pablo Recio, Gonzalo Rodríguez-Ruiz, Jesús Ortega, José Martín* Dept. Ecología Evolutiva, Museo Nacional de Ciencias Naturales, CSIC, José Gutiérrez Abascal 2, 28006 Madrid, Spain. *Corresponding author. E-mail: [email protected]
Submitted on: 2019, 12th February; revised on: 2019, 22nd April; accepted on: 2019, 12th May Editor: Aaron M. Bauer
Abstract. Many field studies of ecology or conservation require individual identification of the animals, and for this, several marking techniques have been developed. However, no specific labeling technique has been tested for fosso- rial reptiles, such as amphisbaenians. We describe the use of Passive Integrated Transponder (PIT) tags as a long- term labeling method of the amphisbaenian Trogonophis wiegmanni. We present the details of the marking procedure and examine the benefits and drawbacks of the technique considering the fossorial environment. After marking many individuals in a long-term field study, we can ensure that the marks were easily applicable and were not lost over a period of at least four years. Moreover, PIT tags did not negatively affect the body condition of amphisbaenians. We conclude that PIT tags are useful for doing field studies of this and similar fossorial species.
Keywords. Amphisbaenians, Trogonophis wiegmanni, PIT-tagging, body condition, fossorial reptiles.
INTRODUCTION Diverse methods of marking individuals have been described for reptiles. Some are intended for short-term Many field studies of ecology, behavior or conserva- studies, such as external painting marks, beads, adhesive tion require individual recognition of the subjects that tapes, elastic bands, metal or plastic discs, buttons, etc. make up a population. Being able to distinguish individu- (Gibbons and Andrews, 2004; Ribeiro and Sousa, 2006; als allows the assessment of diverse ecological traits such Ferner and Plummer, 2016). While others are focused as the size and dynamics of the population, survivorship, on long-term studies, such as toe clipping, scale clipping, movements, home ranges, activity patterns, social inter- shell notching on turtles, heat/freeze branding, photo actions, etc. (reviewed in Plummer and Ferner, 2012; identification based on natural markings, Visible Implant Ferner and Plummer, 2016). For that reason, labeling Elastomer (VIE) tags and/or Passive Integrated Tran- individuals is often necessary, and diverse tagging tech- sponder (PIT) tags (Daniel et al., 2006; Hutchens et al., niques have been developed depending on the species 2008; Ekner et al., 2011; Ferner and Plummer, 2016). and/or the traits that are the object of study. Ideally, these Several groups of reptiles and amphibians, compris- marks should allow a correct identification and be easily ing as much as 20% of the global herpetofauna, or nearly applicable, but without causing suffering to the animals, 3,000 species, are fossorial (Measey, 2006). However, as is and they should last for at least the duration of the entire the case with other fossorial animals, their ecology and field study, but without affecting survival or behavior of conservation status are much less well understood than the marked animals (reviewed in Ferner and Plummer, those of their epigeal relatives (Copley, 2000; Wolters, 2016). 2001; Böhm et al., 2013). This may be explained because
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 102 Pablo Recio et alii of the difficulty of doing field studies of fossorial animals amphisbaenians. More detailed studies would require to indi- (Measey, 2006; Henderson et al., 2016), which includes vidually identify the amphisbaenians that are being examined. difficulties in individually marking these animals, given This is important, not only because of the scientific interest in their burrowing habits. Although several marking tech- understanding the ecology and behavior of amphisbaenians, but because several conservation problems that may potentially niques have been tested in fossorial caecilians (Measey affect their populations have been noted (Martín et al., 2011a, et al., 2001) and Ambystoma salamaders (Connette and 2015, 2017), and a detailed long-term monitoring of these pop- Semlitsch, 2012), to our knowledge, no specific labeling ulations require the ability to individually identify and follow technique has been tested for limbless fossorial reptiles the study subjects. such as amphisbaenians (Henderson et al., 2016). Due to the morphology of most amphisbaenian spe- cies (i.e., elongated body without limbs in most species), Field study and marking procedure it is obviously not possible to use many types of marking methods. Further, given the fossorial habits of amphis- We have carried out field and laboratory studies of T. wieg- manni amphisbaenians on the Chafarinas Islands (Spain) for baenians, most external markings (painting, beads, almost twenty years. This is an archipelago, formed by three adhesive tapes, etc.) may be incompatible with the bur- small islands, located in the southwestern area of the Mediterra- rowing behavior of these animals and will be quickly nean Sea (35°10’N, 02°25’W), 2.5 nautical miles to the north of lost by repeated contact of the body with the soil. There- the Moroccan coast (Ras el Ma, Morocco) and 27 miles to the fore, potential methods that could be used for long-term east of the Spanish city of Melilla. The islands have a dry, warm, marking of amphisbaenians might be restricted to scale Mediterranean climate, and vegetation is dominated by bushy clipping, heat/freeze branding, VIE tags and/or PIT tags plants (Suaeda, Salsola, Lycium and Atriplex) adapted to salin- (Camper and Dixon, 1988; Jemison et al., 1995; Hutch- ity and drought. Trogonophis wiegmanni is very common and is ens et al., 2008; Ferner and Plummer, 2016). Here, we represented by very large populations on these islands (Martín et al., 2011a). describe the use of PIT tags as a labeling method for During the years 2015-2018, we made field campaigns long-term field studies of the checkboard amphisbae- twice a year, during two weeks in spring (March-April) and two nian Trogonophis wiegmanni, Kaup 1830. A PIT tag is a weeks in Autumn (September-October), to capture, mark and microchip with an electromagnetic coil encased in a bio- recapture T. wiegmanni. We delimited three study plots (surface compatible glass cylinder, encoded alphanumerically in area = 0.14 Ha, 0.40 Ha and 0.58 Ha) on different islands, which an unique way, that is implanted in the animal (Gibbons we walked systematically and intensively during the morning and Andrews, 2004). We present here the detailed mark- and afternoon of different days. Amphisbaenians were found ing procedure that we applied to amphisbaenians, exam- by carefully lifting almost all rocks located inside the study ine the potential benefits and drawbacks of the technique, plots. Individuals were captured by hand, measured and imme- diately after marked in the field with PIT tags. We used one of considering the peculiar characteristics of the fossorial the smallest available PIT tags (Biomark MiniHPT8; Biomark, environment, and discuss its utility for doing ecological Inc., Boise, Idaho, USA), with a length of 8.4 mm, 1.4 mm in studies of this and similar fossorial species. diameter and a weigh of 0.03 g. This weigh represents 0.6 % of the mean body mass (i.e., around 5 g) of a typical adult amphis- baenian in our population (Martín et al., 2011c). We gently MATERIALS AND METHODS implanted PIT tags subcutaneously in the upper right side of the body of amphisbaenians (Fig. 1). For this, we made a small Study species puncture at around 3 cm from the snout (mean SVL of adult amphisbaenians is around 14 cm) using a stainless steel needle The checkboard amphisbaenian T. wiegmanni, Kaup 1830 (Biomark N165 needle; length = 5.1 cm, needle diameter = 1.49 is a representative of the family Trogonophidae (Gans, 2005) mm), disinfected with alcohol before and after puncturing each (Fig. 1a) that inhabits arid areas from southwest Morocco to individual, which was fitted to a specially designed syringe style northeast Tunisia (Bons and Geniez, 1996). These amphisbaeni- implanter (Biomark MK165 syringe). We gently lifted the skin ans live all their life buried in the soil, but they are frequently from the underlying muscle and then inserted the transponder found under rocks (Civantos et al., 2003; Martín et al., 2013a). subcutaneously using the implanter. During the insertion of Little research has been carried out on this species, as on other the PIT tag, the needle was maintained parallel to the body to amphisbaenians, but there is now a growing body of informa- ensure that the tag remained under the skin and did not enter tion on aspects such as its thermal biology (López et al., 2002), the coelomic cavity (Fig. 1). The injection site was immedi- microhabitat and soil selection (Civantos et al., 2003; Martín et ately disinfected with alcohol after the implant. According to al., 2013a), reproduction (Bons and Saint Girons, 1963), social Brown (1997), losses of PIT tags may occur immediately after behavior and population structure (Martín et al., 2011b, c) or the implant is done, while the wound is still open. To avoid diet (Martín et al., 2013b; Baeckens et al., 2017). However, all this, incisions may be sealed with medical grade suture glue. these studies have been made by randomly sampling unmarked However, in our case, this was not needed as the incision was Marking fossorial reptiles 103
Fig. 1. An adult amphisbaenian (Trogonophis wiegmanni) as it was found under a stone (left); PIT tag implantation procedure (right). very small and the tags showed no evidence of becoming dis- could have negative physiological side effects for small reptiles placed. Further, at least in lizards, the glue may slow the heal- (Heard, 2001; Chatigny et al., 2017). ing process (Le Galliard et al., 2011). Moreover, the long needle A hand-held portable reader (Biomark 601 Reader) was pushed up the tag under the skin towards the posterior part of used to read the individual unique code of the tag (the tags the animal. Thus, the tag was implanted at least 2 cm posterior have a 134.2 kHz, ISO FDX-B, frequency), The code can be pro- of the small puncture point, which precluded tits loss when the vided either as a hexadecimal or as a decimal number (15 dig- amphisbaenian burrowed forward. All the marking procedure its). In the practice, the four last digits were enough for a reli- could be easily made by a single experienced researcher, hold- able identification of individuals in each study population. The ing the amphisbaenian with one hand and the implanter with reader works in the field with AA rechargeable batteries but it the other. However, the presence of an additional researcher, may be also used in the lab with an AC power supply. who prepared the equipment and took notes, made the process To test the long-term effect of PIT tags in amphisbaenians, easier and quicker, decreasing the manipulation time and dis- we compared the body condition of individuals at first capture, turbance to the animals. when they were untagged, and when they were recaptured one This marking technique is particularly appropriate for year after being implanted with a PIT tag. Body condition was amphisbaenians, as their skin attachment is quite loose and assessed as the residuals of an ordinary least squares linear leaves a subcutaneous space where the pit-tag is inserted. The regression of log-transformed mass (measured with an elec- skin is connected to the axial mass by costocutaneous and ver- tronic balance to the nearest 0.1 g) against log-transformed total tebrocutaneous muscles, that allow the skin to move indepen- length (measured with a metallic ruler to the nearest 1 mm). dently from the body, mainly in rectilinear locomotion (Gans, To ensure that amphisbaenians had empty stomach and intes- 1978; Gasc, 1981; see illustrations in Smalian, 1884), As those tines before being weighed, we gently compressed their vents to muscles are numerous and redundant, the insertion of a strange force the expulsion of feces (used for a study of diet). The small body (or even the damage of some muscle fibers) should not weight of the tag was considered negligible. interfere with the normal locomotion or excavation. Although amphisbaenians obviously “felt” and showed a small aversive response to the puncture with the needle, we RESULTS AND DISCUSSION did not observe any subsequent additional negative behavioral responses (e.g., stress, immobility, forced unnatural movements, In the four years of marking amphisbaenians, we or attempts to remove the tag) (Warwick et al., 2013). Amphis- baenians behaved normally when they were released at their have implanted PIT tags in a mean of 45 + 4 amphisbae- capture points a few minutes after being captured and marked. nians per study plot and campaign (3 plots and 7 cam- The implant procedure very rarely resulted in a small drop of paigns of 15 days each), which so far leads to a grand blood, but in that case the wound was cleaned with alcohol and total of 930 marked individuals in the four years. The bleeding stopped rapidly. We avoided the use of local anesthe- number of individuals found and marked was significant- sia, because the duration of the recovery time from anesthe- ly higher in spring than in autumn for a similar search sia could be much longer than the natural recovery from the effort (F1,19 = 11.82, P = 0.003). implanting procedure. Moreover, the administration of anes- Recapture rate was, however, relatively low; only thesia per se is an additional procedure that requires increas- around 15% of individuals found had already been ing manipulation time and careful control of conditions, and it marked. This is likely attributable to the difficulty of find- 104 Pablo Recio et alii ing the same individual on several occasions in a rela- mating, or risk of predation when PIT tagging different tively short field campaign (i.e., each study plot is sur- species of Liolaemus lizards. Brown (1997) also conclud- veyed only during 3-4 days per campaign) and the high ed that PIT tags did not make any difference in survi- density of amphisbaenians, rather than to the fact that vorship nor body condition of diverse amphibians, and the marking procedure might affect survivorship or that Keck (1994) obtained similar results for growth rates and the tags were lost and we were not be able to identify pre- mobility in several snake species. Nevertheless, females viously marked individuals. In fact, when we captured of the newt Ichthyosaura alpestris laid significantly more an unmarked individual, we always ensured that it had eggs when marked, which seems to be related to a stress no scars at the usual injection point, which may indi- response (Perret and Joly, 2002). Also, measures of corti- cate that it had been marked previously but had no tag costerone in blood have shown that the PIT tag implant- inside. Such scars are typical of marked individuals, but ing procedure can be stressful for small skinks at least 14 they have never been observed in unmarked individuals. days after the implant (Langkilde and Shine, 2006), but Also, we have not noted a decrease of population size, as have no effects on stress five days after in common lizards assessed from the number of individuals usually found (Le Galliard et al., 2011). Our data show that PIT tags do in a working day, which might reflect low survivorship not have a negative long-term impact on the body condi- of marked animals. Moreover, nearby populations on tion of T. wiegmanni (body condition of the same indi- the islands, where we sampled amphisbaenians with- viduals, initial vs. recapture: 0.05 + 0.03 vs. 0.06 + 0.04; out marking them, show similar trends to the marked one-way repeated measures ANOVA: F1,96 = 0.34, P = populations (unpublished data). Therefore, we are confi- 0.56). This lack of change of the body condition is a good dent that the marking procedure is effective and it is not indication of the absence of long-term negative effects of adversely affecting the populations. the PIT tag on health of individuals, as it is known that Several authors have considered that PIT tags are not natural and anthropomorphic alterations of the soil are always permanent (Brown, 1997; Ott and Scott, 1999), reflected in a low body condition of these amphisbaeni- while others claim permanence for more than 20 (Ger- ans (Martín et al., 2015, 2017). mano and Williams, 1993) or 70 years (Ferner and Plum- Another disadvantage associated with PIT tagging mer, 2016). In our study, we have recaptured individuals may be related to the price of the reader and each tran- marked in the first year of the fieldwork after four years sponder (Gibbons and Andrews, 2004). In our case, the and we are confident that the mark will persist during the current price of each PIT tag is $2.58 (they are provided entire life of the amphisbaenian. Gibbons and Andrews in packs of 100 units), the implanters cost $5 each (each (2004) postulated that tag migration may complicate code one is useful for many markings), the needles costs $2 checking when it is not possible to find the tag, and can each (for an optimal functioning, we used a different nee- also lead to health problems when migrating through the dle for every 20 punctures), and the reader costs $595. digestive or urinary systems (Jemison et al., 1995). This These costs may be normally easily assumed by research problem may be greater in fossorial burrowing animals or conservation projects financed by the government or due to the constant friction with the substrate (Measey et other institutions. al., 2001). In our study, although tag migration occurred With respect to the invasiveness of the procedure, in several individual amphisbaenians, in all cases, the tag it has been recommended that it should not be used for had stayed just under the skin and was relocated posteri- individuals smaller than 8 cm (Camper and Dixon, 1988; orly of the injection point, reaching a point close to the Gibbons and Andrews, 2004; Ferner and Plummer, 2016). cloaca in the longest observed migrations. This move- In our study, the small size of the PIT tags allowed us to ment of the tag seems to be along the subcutaneous space mark amphisbaenians as small as 90 mm SVL without typical of amphisbaenians (see above) (Gans, 1978; Gasc, problems, the suitability for being marked depending 1981). We did not detect injuries or health problems (e.g., more on the diameter of the body than on the length. infection, sores, bleeding, low body condition, etc) in These “small” amphisbaenians are second year young any case. Besides, we did not encounbter any problems subadult individuals (Martín et al., 2001c). Only newborn in reading the tag, even in cases where its exact location individuals (SVL<70 mm when they born in autumn) was not easily detected at first sight, probably because the seem unsuitable for marking with these PIT tags, consid- small size of T. wiegmanni allowed us to scan the entire ering the size of the currently available tags. This can be body surface under the reader at the same time. a problem that precludes the study of aspects of popula- On the other hand, long-term effects of PIT tags have tion dynamics, such as growth rates and survivorship of been described for several species. However, Lobos et al. juveniles in their first year. However, given the low move- (2013) did not find significant impacts on growth rates, ment rate of amphisbaenians, is still possible to assess the Marking fossorial reptiles 105 number of individuals born in a population if we control zation for the study was provided by the Spanish National for the geographic location of the newborn individuals Parks Authority and the experimental procedures were found, to ensure that we do not repeat the same individ- supervised by the BioEthical Committee of the Span- ual on different days. We expect that the future develop- ish National Research Council (CSIC). The implanting ment of smaller PIT tags will allow them to be used in all procedure was carried out by researchers holding the individuals. adequate animal experimentation accreditation. Financial In contrast to PIT tags, scale clipping and heat support was provided by the project PGC2018-093592-B- branding may be cheaper (Winne et al., 2006; Ekner et I00 (MCIU/AEI/FEDER, UE) of the Spanish Ministerio al., 2011), whereas VIE tags can be seen without captur- de Ciencia, Innovación y Universidades. ing the subject (Daniel et al., 2006; Hutchens et al., 2008), and none of these techniques are as invasive as PIT tag- ging (Gibbons and Andrews, 2004; Ferner and Plummer, REFERENCES 2016). However, PIT tags offer numerous advantages compared with the other long-term techniques potential- Baeckens, S., García-Roa, R., Martín, J., Ortega, J., ly useful for marking amphisbaenians. PIT tags are per- Huyghe, K., Van Damme, R. (2017): Fossorial and manent and marks are unmistakable, while brands made durophagous: implications of molluscivory for head by scale clipping or heating/freezing procedures may be size and bite capacity in a burrowing worm lizard. J. confounded with natural marks or become unreadable Zool. 301: 193-205. as time passes due to external agents (Winne et al., 2006; Böhm, M., Collen, B. Baillie, J.E.M., Bowles, P., Chanson, Ekner et al., 2011). Similarly, VIE tags might be hard to J. et al. (2013): The conservation status of the world’s detect in darkly pigmented tissues (Hutchens et al., 2008; reptiles. Biol. Cons. 157: 372-385. Petit et al., 2012). Also, PIT tags are able to provide data Bons, J., Geniez, P. (1996): Amphibians and reptiles of even after the death of the marked individual, and/or can Morocco. Asociación Herpetológica Española, Barce- be even reused (Gibbons and Andrews, 2004). Finally, lona. PIT-tag telemetry (i.e., detecting the radiofrequency sig- Bons, J., Saint Girons, H. (1963): Ecologie et cycle sex- nal of the tag at distance; e.g., Connette and Semlitsch, uel des amphisbeniens du Maroc. Bull. Soc. Sci. Nat. 2012; Ousterhout and Semlitsch, 2014) may allow the Phys. Maroc. 43: 117-158. detection and relocation of fossorial animals burrowed Borges-Landáez, P.A., Shine, R. (2003): Influence of toe- underground without physically contacting them. How- clipping on running speed in Eulamprus quoyii, an ever, for this technique a special, and more expensive, australian scincid lizard. J. Herpetol. 37: 592-595. detector with an attached antenna is needed to careful- Brown, L.J. (1997): An evaluation of some marking and ly scan the soil surface, and the detection range for the trapping techniques currently used in the study of smallest 8 mm PIT tags is only 16 cm in depth increas- anuran population dynamics. J. Herpetol. 31: 410-419. ing to 30 cm for a 12 mm PIT tag (Ousterhout and Sem- Camper, J.D., Dixon, J.R. (1988): Evaluation of a micro- litsch, 2014). chip marking system for amphibians and reptiles. In conclusion, although further and more specific Texas Parks and Wildlife Department, Research Pub- studies are needed to test the usefulness and effectiveness lication, 7100-159. Austin, Texas. of PIT tagging in fossorial reptiles, it seems to be a valid Chatigny, F., Kamunde, C., Creighton, C.M., Stevens, E.D. procedure for individual recognition in long-term field (2017): Uses and doses of local anesthetics in fish, studies of amphisbaenians such as Trogonophis wiegman- amphibians, and reptiles. J. Am. Assoc. Lab. Anim. ni, and, therefore, we suggest that it may be also applied Sci. 56: 244-253. to other similar limbless fossorial species. Civantos, E., Martín, J., López, P. (2003): Fossorial life con- strains microhabitat selection of the amphisbaenian Trogonophis wiegmanni. Can. J. Zool. 81: 1839-1844. ACKNOWLEDGMENTS Connette, G.M., Semlitsch, R.D. (2012): Successful use of a passive integrated transponder (PIT) system for We thank Ricardo Montero for useful comments and below-ground detection of plethodontid salamanders. the personal and facilities of the field station of the “ZEC Wildlife Res. 39: 1-6. Islas Chafarinas” (Organismo Autónomo de Parques Copley, I. (2000): Ecology goes underground. Nature 406: Nacionales) for logistical support. We specially thank J.I. 452-454. Montoya, J. Díaz, G. Martínez, A. Sanz, F. López and A. Daniel, J.A., Baker, K.A., Bonine, K.E. (2006): Retention Ruiz for friendship and help in the Islands. Legal authori- rates of surface and implantable methods in the medi- 106 Pablo Recio et alii
terranean house gecko (Hemidactylus turcicus), with descripción de una técnica de implante para especies notes on capture methods and rates of skin shedding. de pequeña y mediana talla. Gayana 77: 26-34. Herpetol. Rev. 37: 319-321. López, P., Civantos, E., Martín, J. (2002): Body tempera- Ekner, A., Sajkowska, Z., Dudek, K., Tryjanowski, P. ture regulation in the amphisbaenian Trogonophis (2011): Medical cautery units as a permanent and wiegmanni. Can. J. Zool. 80: 42-47. non-invasive method of marking lizards. Acta Herpe- Martín, J., Polo-Cavia, N., Gonzalo, A., López, P., Civan- tol. 6: 229-236. tos, E. (2011a): Distribución, abundancia y conser- Elbin, S., Burguer, J. (1994): In my experience…Implant- vación de la culebrilla mora (Trogonophis wiegmanni) able microchips for individual identification in wild en las Islas Chafarinas. Bol. Asoc. Herpetol. Esp. 22: captive populations. Wildl. Soc. Bull. 22: 677-683. 107-112. Ferner, J.W., Plummer M.V. (2016): Marking and measur- Martín, J., Polo-Cavia, N., Gonzalo, A., López, P., Civan- ing reptiles. In: Reptile ecology and conservation. A tos, E. (2011b): Social aggregation behaviour in the handbook of techniques, pp. 45-59. Dodd, C.K., Ed, North African amphisbaenian Trogonophis wiegman- Oxford University Press, Oxford. ni. Afr. J. Herpetol. 60: 171-176. Gans, C. (2005): Checklist and bibliography of the Martín, J., Polo-Cavia, N., Gonzalo, A., López, P., Civ- Amphisbaenia of the world. Bull. Am. Mus. Nat. Hist. antos, E. (2011c): Structure of a population of the 280: 1-130. amphisbaenian Trogonophis wiegmanni in North Afri- Gasc, J.P. (1981): Axial musculature. In: Biology of the ca. Herpetologica 67: 250-257. Reptilia, vol. 11. Morphology F, pp. 355-435. Gans, C., Martín, J., López, P., García, L.V. (2013a): Soil character- Parsons, T.S., Eds. Academic Press, London. istics determine microhabitat selection of the fossorial Germano, D.J., Williams, D.F. (1993): Field evaluation of amphisbaenian Trogonophis wiegmanni. J. Zool. 290: using Passive Integrated Transponder (PIT) tags to 265-272. permanently mark lizards. Herpetol. Rev. 24: 54-56 Martín, J., Ortega, J., López, P., Pérez-Cembranos, A., Gibbons, J.W., Andrews, K.M. (2004): PIT Tagging: sim- Pérez-Mellado, V. (2013b): Fossorial life does not con- ple technology as its best. BioSci. 54: 447-454. strain diet selection in the amphisbaenian Trogonophis Henderson, R.W., Powell, R., Martín, J., López, P. (2016): wiegmanni. J. Zool. 291: 226-233. Sampling techniques for arboreal and fossorial rep- Martín, J., López, P., Gutiérrez, E., García, L.V. (2015): tiles. In: Reptile ecology and conservation. A hand- Natural and anthropogenic alterations of the soil book of techniques, pp. 139-153. Dodd, C.K., Ed, affect body condition of the fossorial amphisbaenian Oxford University Press, Oxford. Trogonophis wiegamnni in North Africa. J. Arid Envi- Heard, D.J. (2001): Reptile anesthesia. Veter. Clin. N. ron. 122: 30-36. Amer.: Exot. Anim. Pract. 4: 83-117. Martín, J., Gutiérrez, E., García, L.V. (2017): Altera- Hutchens, S.J., Deperno, C.S., Matthews, C.E., Pollock, tion effects of ornamental whitewashing of rocks on K.H., Woodward, D.K. (2008): Visible Implant Fluo- the soil properties and body condition of fossorial rescent Elastomer: a reliable marking alternative for amphisbaenians that live under them. Herpetol. Con- snakes. Herpetol. Rev. 39: 301-303. serv. Biol. 12: 367-372. Jemison, S.C., Bishop, L.A., May, P.G., Farrell, T.M. Measey, G.J. (2006): Surveying biodiversity of soil her- (1995): The impact of PIT-tags on growth and move- petofauna: towards a standard quantitative methodol- ment of the rattlesnake Sistrurus miliarus. J. Herpetol. ogy. Eur. J. Soil Biol. 42: S103-S110. 29: 129-132. Measey, G.J., Gower, D.J., Oommen, O.V., Wilkinson. Keck, M.B. (1994): Test for detrimental effects of PIT tags M. 2001. Permanent marking of a fossorial caecilian, in neonatal snakes. Copeia 1994: 226-228. Gegeneophis ramaswamii (Amphibia: Gymnophiona: Langkilde, T., Shine, R. (2006): How much stress do Caeciildae). J. South Asian Nat. Hist. 5: 141-147. researchers inflict on their study animals? A case Ott, J.A., Scott, D.E. (1999): Effect of toe-clipping and study using a scincid lizard, Eulamprus heatwolei. J. PIT-tagging on growth and survival in metamorphic Exp. Biol. 209: 1035-1043. Ambystoma opacum. J. Herpetol. 33: 344-348. Le Galliard, J., Paquet, M., Pantelic, Z., Pret, S. (2011): Ousterhout, B.H., Semlitsch, R.D. (2014): Measuring ter- Effects of miniature transponders on physiologi- restrial movement behavior using passive integrated cal stress, locomotor activity, growth and survival in transponder (PIT) tags: effects of tag size on detec- small lizards. Amphib.-Rept. 32: 177-183. tion, movement, survival, and growth. Behav. Ecol. Lobos, G., Méndez, C., Alzamora, A. (2013): Utilización Sociobiol. 68: 343-350. de marcas electrónicas “PIT tags” en Liolaemus y Perret, N., Joly, P. (2002): Impacts of tattooing and PIT- Marking fossorial reptiles 107
tagging on survival and fecundity in the alpine newt (Triturus alpestris). Herpetologica 58: 131-138. Petit, S., Waudby, H.P., Walker, A.T., Zanker, R., Rau, G. (2012): A non-mutilating technique for marking small wild mammals and reptiles. Austr. J. Zool. 60: 64-71. Plummer, M.V., Ferner, J.W. (2012): Marking reptiles. In: Reptile biodiversity: standard methods for inventory and monitoring, pp. 143-150. McDiarmid, R.W., Fos- ter, M.S., Guyer, C. et al., eds., University of California Press, Berkeley, California. Ribeiro, L.B., Sousa, B.M. (2006): Elastic hair bands: an effective marking technique for lizards in mark-recap- ture studies. Herpetol. Rev. 37: 434-435. Smalian, C. (1884): Beitrage zur Anatomie der Amphis- baeniden. Z. Wiss. Zool. 42: 126-202. Warwick, C., Arena, P.C., Lindley, S., Jessop, M., Steed- man, C. (2013): Assessing reptile welfare using behav- ioural criteria. In Pract. 35: 123-131. Winne, C.T., Willson, J.D., Andrews, K.M., Reed, R.N. (2006): Efficacy of marking snakes with disposable medical cautery units. Herpetol. Rev. 37: 25-54. Wolters, V. (2001): Biodiversity of soil animals and its function. Eur. J. Soil Biol. 37: 221-227.
Acta Herpetologica 14(2): 109-115, 2019 DOI: 10.13128/a_h-7748
Occurrence of Batrachochytrium dendrobatidis in the Tensift region, with comments on its spreading in Morocco
Redouane Ait El Cadi1, El-Mustapha Laghzaoui1, Angelica Crottini2, Tahar Slimani1, Jaime Bosch3,4, El Hassan EL Mouden1,* 1 Laboratory of Biodiversity and Ecosystem Dynamic, Faculty of Sciences, Semlalia, Cadi Ayyad University, Marrakech, Morocco. *Cor- responding author. Email: [email protected] 2 CIBIO Centro de Investigação em Biodiversidade e Recursos Genéticos, Vairão, Portugal 3 Museo Nacional de Ciencias Naturales, CSIC, Madrid, Spain 4 UMIB Research Unit of Biodiversity (CSIC, UO, PA), Universidad de Oviedo, Campus de Mieres, Spain
Submitted on: 2019, 27th February; revised on: 2019, 4th September; accepted on: 2019, 17th September Editor: Adriana Bellati
Abstract. The chytrid fungus Batrachochytrium dendrobatidis (Bd) is a generalist pathogen that affects many amphib- ian species and is responsible of chytridiomycosis onset, considered as the main causes of species extinctions and populations declines worldwide. The chytrid fungal pathogen has been first described in North Africa in 2011. The present work reported the first survey onBd prevalence and intensity in the Tensift region of Morocco. The survey has been conducted on 11 different localities by collecting skin swabs and tissue samples of 97 individuals. Using a quan- titative Polymerase Chain Reaction (qPCR) protocol, low-intensity of Bd infection has been detected in the area of study. In fact, the chytrid fungal pathogen has been identified in 10 individuals distributed in six of the 11 sites inves- tigated, placing the 95% confidence interval for overall prevalence at 5.5-19.6%. The survey confirmed the occurrence of Bd at both high and low altitude localities, on four species out of seven known to inhabit the region and added two additional species (Pelophylax saharicus and Sclerophrys mauritanica) to the list of Bd susceptible amphibians in Morocco. The present records extended Bd distribution more than 400 km in the South of Morocco, indicating that the chytrid fungal pathogen is more widespread in the country than previously thought.
Keywords. Batrachochytrium dendrobatidis, amphibians, Tensift, prevalence, intensity.
INTRODUCTION the first wildlife fungal pathogen to have caused wide- spread species extinctions (Skerratt et al., 2007), and Habitat degradation and overexploitation are the has been implicated in rapid population declines and main large-scale factors causing loss of biodiversity extinction of several amphibians (Retallick et al., 2004; worldwide (Hoffmann et al., 2010). While these two fac- Schloegel et al., 2006; Skerratt et al., 2007; Kolby and tors are to blame for the rapid declines of many verte- Daszak, 2016; Lips, 2018). It has been detected in about brate species, amphibians continue to decline also in 48% of tested amphibian species (Olson et al., 2013), undisturbed habitats, due essentially to chytridiomyco- and is recently known to be rapidly expanding its glob- sis, an emerging infectious skin disease caused by the al range of distribution (Fisher et al., 2009; O’Hanlon et chytrid fungus Batrachochytrium dendrobatidis (Bd) al., 2018). In North Africa, Bd has been detected for the (Berger, 1998; Longcore and Pessier, 1999; Bosch and first time in Morocco at three (out of 51) tested locali- Martínez-solano, 2006; Lips, 2016). Bd is considered as ties, and was found in only three species (Discoglos-
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 110 Ait El Cadi Redouane et alii sus scovazzi, Hyla meridionalis and Pelobates varaldii), with a total prevalence of 6% (El Mouden et al., 2011). The sites where Bd has been detected are in the north- ern part of the country (Tingitana peninsula and its sur- roundings) near South of Spain. El Mouden et al. (2011) explained this presence through a possible arrival from Spain where chytrid fungus is largely distributed, and it’s responsible for severe decline of amphibian popula- tions in mountain areas (Bosch et al., 2001; Bosch et al., 2007; Walker et al., 2010; Lips, 2016). These findings suggest further investigations in other areas were needed to collect more information on the potential extent of the spread of this fungal pathogen in Morocco, and to elucidate its potential source. Morocco is home to many endangered amphibian populations that are already threatened by habitat alteration and destruction, pollu- tion, and climatic changes (Fahd et al., 2015; Ben Hass- ine and Nouira, 2012; Escoriza and Ben Hassine, 2017). The aim of this study is to evaluate the chytrid fungal presence and prevalence using qPCR test for amphib- ians naturally occurred in Tensift region, an area located around 400 km South from the localities where Bd was detected for the first time in 2011. This region is char- acterized by a high diversity of habitats from wetlands (High Atlas Mountains) to arid environments (Jbilets), with seven amphibians species known to inhabit the area, two of which are Moroccan endemics (Bons and Geniez, 1996; Beukema et al., 2013). Fig. 1. Map of the Northern Morocco showing the result of Batra- chochytrium dendrobatidis (Bd) sampling sites realized during the MATERIALS AND METHODS present study (circles); dark circles indicate populations with Bd- positive samples and open circles indicate populations with Bd- Surveys were conducted between October 2013 and Octo- negative samples. Within Bd positive sites, pie charts indicate the ber 2017 in the Tensift region (central Morocco). The study proportion of infected individuals in black and uninfected individu- area covers about 20,000 km2, surrounded by the southern als in white. Squares indicate infected populations as reported by El Mouden et al. (2011). crest of the High Atlas Mountains (with an altitude up to 4000 m a.s.l.) and the northern Jbilet hills (separated by the Haouz plain), while the coastline of Essaouira extends in the East (Fig. 1). The climate of the region is characterized by strong annual Daudin, 1803 (Near threatened in Morocco), Barbarophryne variability, with mean temperatures ranging between a maxi- brongersmai (Hoogmoed, 1972) (endemic to Morocco and Alge- mum of 37.7 °C and a minimum of 4.9 °C. The rainfall is gen- ria), Pelophylax saharicus (Boulenger, 1913) (endemic to North erally irregular, with occasional prolonged droughts. The mean Africa) and Discoglossus scovazzii Camerano, 1878 (endemic to annual rainfall varies from 800 mm in the mountain to 190 mm Morocco). in the plain, with significant snowfall between December and Amphibians were searched opportunistically in 11 sites March at high elevation (up to 2000 m a.s.l.) (Alaoui Haroni et (Table 1; Fig. 1), during the rainy season (when detectabil- al., 2009). The semi-arid environment dominates throughout ity of amphibians is higher), during nocturnal and occasional the region, while the sub-humid zones appear at high altitude diurnal surveys. We extensively searched in all water bod- (up to 1500 m a.s.l.). The Tensift region is characterized by a ies available in the area. After capture, animals were swabbed complex landscape and topography, including escarpments, (using Medical Wire Sterile Dryswab™ - MW100) and released floodplains, and a great variety of aquatic ecosystems (rivers, immediately after sampling to the place of capture. To prevent permanent ponds and a large network of Mediterranean tempo- the transmission of diseases by animals, each individual was rary ponds). This diversity of water ecosystems provides a wide handled with disposable gloves. In total, 86 individuals were availability of breeding sites for seven native amphibian species: swabbed. Swabs were then air-dried and stored at cool tem- Hyla meridionalis Boettger, 1874, Bufotes boulengeri (Lataste, perature (4 °C) until processing. Tissues samples have been 1879), Sclerophrys mauritanica (Schlegel, 1841), Bufo spinosus collected from 11 specimens that were found dead in two sites Survey of Batrachochytrium dendrobatidis in Morocco 111
Table 1. Surveyed localities of the Tensift region (Morocco) with date and approximate coordinates and altitude (meters a.s.l.) of each site. Species names are abbreviated as follows: Sm, Sclerophrys mauritanica; Bb, Bufotes boulengeri; Br, Barbarophryne brongersmai; Ds, Discoglos- sus scovazzi; Hm, Hyla meridionalis; Ps, Pelophylax saharicus. Life history stage (LHS): Ad, adult; Juv, juvenile. P/S: number of individuals testing positive / Number of analyzed individuals. Bd load in genomic equivalents of zoospores. (*) Indicate individual found dead in the field and tested positive for Bd.
Date Locality Coordinates Altitude Species LHS P/S Bd load Oct 2013 Sidi Bouathman 31.9 -7.92 ~ 517 Ps Ad 0/1 Nov 2013 Ijoukak N’fis 31.059 -8.164 ~ 1042 Ps Juv 0/1 Ad 1/3 26.8 Dec 2013 Sidi Bouathman 31.9 -7.92 ~ 517 Ps Ad 0/3 Dec 2013 Dam d’Ouled Abbas 31.966 -8.447 ~ 468 Sm* Ad 1/7 0.3 Dec 2013 Dam-Oukaimeden 31.208 -7.851 ~ 2623 Ps Ad 0/7 Dec 2013 Tighedouine 1 31.4 -7.532 ~ 1145 Ps Ad 0/6 Avr 2014 Tighedouine 2 31.423 -7.524 ~ 1066 Ps Ad 1/4 0.3 Hm Ad 1/7 0.4 Avr 2014 Ijoukak Tizgui 30.971 -8.125 ~ 1142 Ds Ad 0/2 Bb Ad 0/1 Ps Ad 0/5 Dec 2014 El Gantour 1 32.189 -8.335 ~ 402 Ps Ad 0/5 El Gantour 2 Dec 2014 32.187 -8.329 ~ 398 Ds Ad 0/2 Ps Juv 0/1 Ps Ad 1/8 1.3 Jan 2015 Jaidate 31.87 -7.79 ~ 623 Bb Ad 0/9 Br Ad 0/3 Ps Ad 0/2 Avr 2015 Tighedouine 2 31.423 -7.524 ~ 1066 Ps Ad 1/1 1.7 Jun 2016 Tifergine-Oukaimeden 31.207 -7.84 ~ 2565 Ds Ad 1/13 1.3 Oct 2017 Dam-Oukaimeden 31.208 -7.851 ~ 2623 Ps* Ad 3/6 0.6, 3.0, 0.3
(Dam-Oukaimden and Dam d’Ouled Abbas). These 11 tissue RESULTS samples were preserved in ethanol (Brem et al., 2007; Hyatt et al., 2007). In total, 97 amphibian specimens were sampled The numbers of screened and testing positive indi- belonging to four families and six species. Of these, 56 indi- viduals from each population and sampling event, along viduals were sampled in the high Atlas Mountains, 16 in the with geographic information on the sampling site and arid area of Jbilets, 23 in the El Gantour region and two in the Haouz plain (Table 1; Fig. 1). year of capture are reported in Table 1. The results show DNA was extracted from both swabs and tissue samples that 10 out of 97 screened individuals were Bd infected using PrepMan Ultra reagent and extractions were diluted (10.3%), corresponding to a 95% confidence interval of 1:10 before real-time PCR amplification, performed in dupli- overall Bd prevalence of 5.5-19.6%. Across the four spe- cate with a CFX96 thermocycler (Bio-Rad), following Boyle et cies that tested positive, the infection prevalence was al. (2004). Each 96-well assay plate included samples, a nega- 11.9% (95% CI: 5.5-19.6%). This study confirmed Bd tive control and standards of 100, 10, 1, and 0.1 Bd zoospore occurrence in both the northern and southern parts of genome equivalents in duplicate. Samples were considered the Tensift watershed at elevations between 468 and 2625 positives when both replicates were ≥ 0.1 and the amplification m a.s.l (Fig. 1). curves had the typical sigmoidal shape. When only one replicate from any sample amplified, we ran this sample a third time. If In the study area, Bd was detected in six out of 11 the third amplification did not result in an amplification profile, investigated sites (Table 1), with prevalence values rang- we considered sample as negative for infection. Samples that ing between 0 and 25%, and lower GE (genomic equiv- showed signs of inhibition (non-sigmoidal amplification) were alent) values that varied between 0.3 and 26.8. These further diluted to 1:100 and re-analyzed. If signs of inhibition results indicate a significant presence ofBd in the Ten- remained, the samples were excluded. sift region, with low infection prevalence and intensi- ties across all sites that tested positive for Bd. However, 112 Ait El Cadi Redouane et alii
Table 2. Bd prevalence with 95% Clopper-Pearson binomial confi- DISCUSSION dence intervals for individuals grouped by species and elevation. In our study area, four out of six amphibian species Bd sample have been tested as Bd positive, although in comparison Group prevalence 95% CI size (%) with other regions in Marocco, they have lower infection rate. Lower GE values are probably the result of unfa- Family/Species vorable conditions, such as changes in the abiotic envi- Ranidae ronment (Ron, 2005; Thorpe et al., 2018). It has been Pelophylax saharicus 53 13.2 5.6 - 25.8% reported that Bd is temperature sensitive, with optimal Alytidae growth ranging between 17 and 25 °C (Piotrowski et Discoglossus scovazzi 17 5.9 0.1 - 28.7% al., 2004; Pounds et al., 2006), with higher temperature Bufonidae (more than 30 °C) reported as unfavorable for its devel- Bufotes boulengeri 10 0 0 - 30.8 % opment (Watve, 2013). Previous studies have shown that Sclerophrys mauritanica 7 14.3 0.4 - 57.9% lower temperature regime resulted in extended zoospore Barbarophryne brongersmai 3 0 0 - 70.8% longevity and in such conditions zoospores numbers Hylidae in water bodies could be expected to be greater than in Hyla meridionalis 7 14.3 0.4 - 57.9% Elevation (for all species) warmer water (Voyles et al., 2012; Thorpe et al., 2018). Less than 700 m 41 4.9 0.6 - 16.2% The higher temperatures (more than 30 °C) recorded in More than 700 m 56 14.3 6.1 - 25.4% the study area expected to have a negative impact on the development of this fungal pathogen. Additionally, opti- mum rainfall for Bd development has been reported to range between 1500 and 2500 mm/year (Thorpe et al., in October 2017, 50% prevalence was recorded in Dam- 2018), which are much higher values of the one observed Oukaimden, with three testing positive samples out of in the Tensift region. Only, Oukaimden approach these the six screened individuals, which represents the maxi- rainfall values, but in general remains below to optimum mum prevalence obtained in this study. parameters for the development of Bd. It seems that Bd prevalence can vary among spe- In Morocco, Bd surveys were previously conducted cies and elevations (Table 2). Consequently, 95% confi- only in the northern Tingitana area. El Mouden et al. dence intervals for Bd prevalence for individuals grouped (2011) reported Bd occurrence in three sampling sites. according to these two parameters were determined. All Through this study, Bd occurrence has been reported in confidence intervals were overlapping. Thus, no signifi- six additional Moroccan localities. Considering all these cant taxonomic and altitudinal difference in Bd preva- data, 14.5% of the screened Moroccan localities tested lence has been detected given our sample sizes. Howev- positive to Bd. The finding that Bd was recorded in two er, the obtained results tend towards a more significant separated areas suggested a possible wider distribution prevalence at higher elevations (14.3% vs 4.9%). Likewise, across the country. Similarly, Bd was detected in different species infection rate varied between 0 and 14.3%. The sites ranging from 400 to over 2600 m a.s.l, indicating a higher percentage was observed in three species. Bd was probable wide distribution also along the altitudinal gra- found in seven out of 53 P. saharicus across five sampling dient. Sample size is still limited and should be extended sites. In the Dam-Oukaimden site, where a large num- to have a more robust overview of the pattern of Bd pres- bers of dead P. saharicus were found in October 2017, we ence across Morocco. detected Bd on three out of six tested individuals (95% The report on the occurrence of Bd in the Tingitana CI: 11.8-88.2%), which represents the highest preva- peninsula (North of Morocco) was potentially explained lence recorded by species and by site during this study. by the extensive commercial trade of products and ani- For both H. meridionalis (Tighedouine region; April mals across the straits of Gibraltar with a possible intro- 2014) and S. mauritanica (Culinary dam of Ouled Abbas; duction of Bd from Spain (El Mouden et al., 2011). This December 2013), Bd has been detected in one out of hypothesis is in agreement with the explanations of seven individuals, corresponding to a confidence interval O’Hanlon et al. (2018) who found Bd spreading out of for the two species of 0.40-57.9%. During our investiga- Asia and dated its emergence on the early 20th century, tion at Ouled Abbas dam, numerous dead specimens of coinciding with the international expansion of commer- S. mauritanica were observed in the water. Seven tissue cial trade in amphibians for exotic pet, medical, and food samples were collected and screened. The results showed purposes. The presence of Bd in the Tensift region can that one sample tested positive for Bd. be explained by a dissemination process from the North Survey of Batrachochytrium dendrobatidis in Morocco 113 of the country to the South, but also by a possible inde- the Hassan II Academy of Sciences and Technology. The pendent episode of introduction through other commer- Portuguese National Funds through FCT (Foundation for cial routes, such as airports and harbors. The spreading of Science and Technology) support the Investigator FCT Bd through the country can be carried out by potential grant to AC (IF/00209/2014). natural vectors, including waterfowl on their feathers or feet (Johnson and Speare, 2005; Garmyn et al., 2012; Bur- rowes and De La Riva, 2017), water (Johnson and Speare, REFERENCES 2003), and non-susceptible to chytridiomycosis amphib- ians acting as carrier (Kolby et al., 2015). Alaoui Haroni, S., Alifriqui, M., Simonneaux, V. (2009): In Morocco, as in most regions around the world, the Recent dynamics of the wet pastures at Oukaimeden detected Bd belongs to the Global Panzootic Lineage (J. plateau (High Atlas). Biodivers. Conserv. 18: 167-189. Bosch, unpublished data). This is a highly virulent and Ben Hassine, J., Nouira, S. (2012): Répartition highly transmissible chytrid fungus, which is currently géographique et affinités écologiques des Amphibiens infecting more that 700 amphibian species worldwide de Tunisie. Rev. Écol. (Terre Vie) 67: 437-457. (Olson and Ronnenberg, 2014; Lips, 2016). We detect- Berger, L., Speare, R., Daszak, P., Green, D.E., Cunning- ed a mass mortality of amphibians in two localities (S. ham, A.A., Goggin, C.L., Slocombe, R., Ragan, M.A., mauritanica in Dam d’Ouled Abbas and P. saharicus in Hyatt, A.D., McDonald, K.R., Hines, H.B., Lips, K.R., Dam-Oukaimden) and some of the sampled individuals Marantelli, G., Parkes, H. (1998): Chytridiomycosis tested positive for Bd infection. However, in the absence causes amphibian mortality associated with population of a detailed study, we have no evidence of Bd-associated declines in the rain forests of Australia and Central amphibian mortality especially because mortality of indi- America, Proc. Natl. Acad. Sci. USA 95: 9031-9036. viduals generally occurred with higher infection rates Beukema, W., De Pous, P., Donaire-Barroso, D., Bogaerts, than those found in the present study. Other biotic and S., Garcia-Porta, J., Escoriza, D., Arribas, J.O., El abiotic factors can be the cause of the witnessed mass Mouden, E.H., Carranza, S. (2013): Review of the sys- mortality (e.g., Croteau et al., 2008; Hayes et al., 2010; tematics, distribution, biogeography and natural his- Relyea et al., 2012; Whittaker et al., 2013; Budzik et al., tory of Moroccan amphibians. Zootaxa 3661: 1-60. 2014; De Wijer et al., 2018). However, it is worth noting Bons, J., Geniez, P. (1996): Amphibiens et reptiles that the two new species found infected by Bd (P. s aha - du Maroc (Sahara Occidental compris): atlas ricus and S. mauritanica) have a wide distribution and biogeographique. Asociación Herpetológica Española, abundance in the southern Mediterranean region (Bons Barcelona. and Geniez, 1996; Schleich et al., 1996; Mateo et al., Bosch, J., Carrascal, J.M., Durán, L., Walker, S., Fish- 2013), which can make Bd dissemination faster. er, M.C. (2007): Climate change and outbreaks of A systematic survey to determine the impact of Bd amphibian chytridiomycosis in a montane area of occurrence in Morocco is crucial to better characterize Central Spain; is there a link? Proc. R. Soc. B 274: its impact on the amphibian’s population dynamic. In 253-260. the mean time, other Bd positive sites located in Tens- Bosch, J., Martínez-Solano, I. (2006): Chytrid fungus fit region have been recorded and helped extending the infection related to unusual mortalities of Salaman- known Bd occurrence area in Morocco, both at the high dra salamandra and Bufo bufo in the Peñalara Natural and the low altitudes (486-2625 m a.s.l.). In addition, Park , Spain. Oryx 40: 84-89. having detected Bd at multiple sites and in two new Bosch, J., Martínez-Solano, I., García-París, M. (2001): amphibian species contributes to the growing knowledge Evidence of a chytrid fungus infection involved in the on the global pattern of Bd distribution in Morocco and decline of the common midwife toad (Alytes obstetri- North Africa. cans) in protected areas of central Spain. Biol. Con- serv. 97: 331-337. Boyle, D.G., Boyle, D.B., Olsen, V., Morgan, J.A.T., ACKNOWLEDGMENTS Hyatt, A. D. (2004): Rapid quantitative detection of chytridiomycosis (Batrachochytrium dendrobatidis) We would like to thank “Haut Commissariat aux in amphibian samples using real-time Taqman PCR Eaux et Forêts et à la Lutte Contre la Désertication assay. Dis. Aquat. Org. 60: 141-148. (HCEFLCD)” for the Permit to work on the field. C. Brem, F., Mendelson, J.R. III, Lips, K.R. (2007): Field- Monsalve-Caraño helped with qPCR analyses. This sampling protocol for Batrachochytrium dendroba- research was supported by Project [ICGVSA] founded by tidis from living amphibians, using alcohol preserved 114 Ait El Cadi Redouane et alii
swabs. Version 1.0. Electronic document accessible at H., Phillott, R., Campbell, R., Marantelli, G., Gleason, http://www.amphibians.org Conservation Internation- F., Colling, A. (2007): Diagnostic assays and sampling al, Arlington. protocols for the detection of Batrachochytrium dend- Budzik, K.A., Budzik, K.M., Kukiełka, P., Łaptaś, A., Bres, robatidis. Dis. Aquat. Org. 73: 175-192. E.E. (2014): Water quality of urban water bodies – a Kolby, J.E., Daszak, P. (2016): The emerging amphibian threat for amphibians? Ecol. Q. 19: 57-65. fungal disease, chytridiomycosis: a key example of the Burrowes, P.A., De La Riva, I. (2017): Detection of the global phenomenon of wildlife emerging infectious amphibian chytrid fungus Batrachochytrium dend- diseases, Microbiol. Spectr. 4: 1-17. robatidis in museum specimens of Andean aquatic Kolby, J.E, Ramirez, S.D., Berger, L., Richards-Hrdlicka, birds: Implications for pathogen dispersal. J. Wildl. K.L., Jocque, M., Skerratt, L.F. (2015): Terrestrial dis- Dis. 53: 349-55. persal and potential environmental transmission of Croteau, M.C., Hogan, N., Gibson, J.C., Lean, D., the amphibian chytrid fungus (Batrachochytrium den- Trudeau, V.L. (2008): Toxicological threats to amphib- drobatidis). PLoS ONE 10: e0125386. ians and reptiles in urban environments. In: Urban Johnson, M.L., Speare, R. (2003): Survival of Batra- Herpetology, pp. 197-210. Mitchell, J.C., Brown, chochytrium dendrobatidis in water: quarantine and R.E.J., Bartholomew, B., Eds, Society for the Study of disease control implications. Emerg. Infect. Dis. 9: Amphibians and Reptiles, Salt Lake City, Utah. 922-925. De Wijer, P., Watt, P.J., Oldham, R.S. (2018): Amphibian Johnson, M.L., Speare, R. (2005): Possible modes of dis- decline and aquatic pollution: effects of nitrogenous semination of the amphibian chytrid Batrachochytri- fertiliser on survival and development of larvae of the um dendrobatidis in the environment. Dis. Aquat. frog Rana temporaria. Appl. Herpetol. 1: 3-12. Organ. 65: 181-186. El Mouden, E.H., Slimani, T., Donaire-Barroso, D., Lips, K.R. (2016): Overview of chytrid emergence and Fernández-Beaskoetxea, S., Fisher, M. C., Bosch, impacts on amphibians. Phil. Trans. R. Soc. B 371: J. (2011): First record of the chytrid fungus Batra- 20150465. chochytrium dendrobatidis in North Africa. Herpeto- Lips, K.R. (2018): Witnessing extinction in real time. logical Review 42: 71-75. PLoS Biol 16: e2003080. Escoriza, D., Ben Hassine, J. (2017): Diversity of guilds of Longcore, J., Pessier, A. (1999): Batrachochytrium den- amphibian larvae in north-western Africa. PLoS ONE drobatidis gen. et sp. nov., a chytrid pathogenic to 12: 1-18. amphibians. Mycologia 91: 219-227. Fahd, S., Mediani, M., Ohler, A.O., Denys, C.D., Santos, Mateo, J., Geniez, P., Pether, J. (2013): Diversity and con- X. (2015): Diversité et conservation de la faune batra- servation of Algerian amphibian assemblages. Chapter chologique du bassin versant d’oued Laou (Rif, nord- 26. Basic Appl. Herpetol. 27: 51-83 ouest du Maroc), Trav. Inst. Sci. Sér. Gén. 8: 69-84. O’Hanlon, S.J., Rieux, A., Farrer, R.A., Rosa, G.M., Wald- Fisher, M.C., Garner, T.W.J., Walker, S.F. (2009): Global man, B., et al. (2018): Recent asian origin of chytrid emergence of Batrachochytrium dendrobatidis and fungi causing global amphibian declines. Science 360: amphibian chytridiomycosis in space, time, and host. 621-627. Annu. Rev. Microbiol. 63: 291-310. Olson, D.H., Aanensen, D.M., Ronnenberg, K.L., Powell, Garmyn, A., Van Rooij, P., Pasmans, F., Hellebuyck, T., C.I., Walker, S.F., et al. (2013): Mapping the global Van Den Broeck, W., Haesebrouck, F., Martel, A. emergence of Batrachochytrium dendrobatidis, the (2012): Waterfowl: Potential environmental reservoirs amphibian chytrid fungus. PloS ONE 8: 1-13. of the chytrid fungus Batrachochytrium dendrobatidis. Olson, D.H., Ronnenberg, K.L. (2014): Mapping Project: PLoS ONE 7: e35038. 2014 Update. FrogLog 22: 17-21. Hayes, T.B., Falso, P., Gallipeau, S., Stice, M. (2010): The Piotrowski, J.S., Annis, S.L., Longcore, J.E. (2004): Physi- cause of global amphibian declines: a developmental ology of Batrachochytrium dendrobatidis, a chytrid endocrinologist ’s perspective. J. Exp. Biol. 213: 921- pathogen of amphibians. Mycologia 96: 9-15. 933. Pounds, J.A., Bustamante, M.R., Coloma, L.A., Consue- Hoffmann, M., Hilton-Taylor, C., Angulo, A., Bohm, M., gra, J.A., Fogden, M.P.L, Foster, P.N., La Marca, E., Brooks, T.M., et al. (2010): The impact of conserva- Masters, K.L., Merino-Viteri, A., Puschendorf, R., tion on the status of the world ’s vertebrates. Science Ron, S.R., Sanchez-Azofeifa, G.A., Still, C.J., Young, 330: 1503-1509. B.E. (2006): Widespread amphibian extinctions from Hyatt, A.D., Boyle, D.G., Olsen, V., Boyle, D.B., Berger, L., epidemic disease driven by global warming. Nature Obendorf, D., Dalton, A., Kriger, K., Hero, M., Hines, 439: 161-167. Survey of Batrachochytrium dendrobatidis in Morocco 115
Relyea, R.A., Tejedo, M., Torralva, M. (2012): Under- standing of the impact of chemicals on amphibian : a meta-analytic review. Ecol. Evol. 2: 1382-1397. Retallick, R.W.R., Mccallum, H., Speare, R. (2004): Endemic infection of the amphibian chytrid fungus in a frog community post-decline, PLoS Biol. 2: e351. Ron, S.R. (2005): Predicting the distribution of the amphibian pathogen Batrachochytrium dendrobatidis in the New World. Biotropica 37: 209-221. Schloegel, L.M., Hero, J., Berger, L., Speare, R., Mcdonald, K., Daszak, P. (2006): The decline of the sharp-snout- ed day frog (Taudactylus acutirostris): the first docu- mented case of extinction by infection in a free-rang- ing wildlife species? EcoHealth 3: 35-40. Skerratt, L.F., Berger, L., Speare, R., Cashins, S., Mcdon- ald, K.R., Phillott, A.D., Hines, H.B., Kenyon, N. (2007): Spread of chytridiomycosis has caused the rapid global decline and extinction of frogs. Eco- Health 4: 125-134. Thorpe, C.J., Lewis, T.R., Fisher, M.C., Wierzbicki, C.J., Kulkarni, S., Pryce, D., Davies, L., Watve, A., Knight, M.E. (2018): Climate structuring of Batrachochytrium dendrobatidis infection in the threatened amphibians of the northern Western Ghats, India. R. Soc. Open Sci. 5: 180211. Voyles, J., Johnson, L.R., Briggs, C.J., Cashins, S.D., Alford, R.A., Berger, L., Skerratt, L.F., Speare, R., Rosenblum, E.B. (2012): Temperature alters reproduc- tive life history patterns in Batrachochytrium dendro- batidis, a lethal pathogen associated with the global loss of amphibians. Ecol. Evol. 2: 2241-2249. Walker, S.F., Bosch, J., Gomez, V., Trenton, W.J., Andrew, A., Schmeller, D.S., Ninyerola, M., Henk, D.A., Ginestet, C., Arthur P.A., Fisher, M.C. (2010): Fac- tors driving pathogenicity vs. prevalence of amphibian panzootic chytridiomycosis in Iberia. Ecol. Lett. 13: 372-382. Watve, A. (2013): Status review of rocky plateaus in the northern Western Ghats and Konkan region of Maha- rashtra, India with recommendations for conservation and management. J. Threat. Taxa 5: 3935-3962. Whittaker, K., Koo M.S., Wake D.B., Vredenburg V.T. (2013): Global Declines of Amphibians. In: Encyclo- pedia of Biodiversity, second edition, pp. 691-699. Levin S.A., Ed, Academic Press, Cambridge.
Acta Herpetologica 14(2): 117-122, 2019 DOI: 10.13128/a_h-7749
Ontogenetic and interspecific variation in skull morphology of two closely related species of toad, Bufo bufo and B. spinosus (Anura: Bufonidae)
Giovanni Sanna Naturalis Biodiversity Center, P.O. Box 9517, 2300 RA Leiden, The Netherlands Department of Biological, Geological, and Environmental Sciences, University of Bologna, Via Selmi 3, 40126 Bologna, Italy Marine Research Department, Senckenberg am Meer, Südstrand 40, 26382 Wilhelmshaven, Germany E-mail: [email protected]
Submitted on: 2019, 31st January; Revised on: 2019, 6th June; Accepted on: 2019, 5th August Editor: Marcello Mezzasalma
Abstract. Using micro-CT and 3D landmark-based geometric morphometrics, I investigated postmetamorphic shape variation in the skull of Bufo bufo and Bufo spinosus, two widespread European toad species with small phenotyp- ic differences. Two ontogenetic series were compared, for a total of 58 individuals. They exhibited similar allometric growth patterns, characterised by cranial widening with relative shortening and dorsoventral compression. However, some interspecific shape divergence was observed, particularly among adults: a relatively shorter skull and a more dorsally extended snout distinguished B. spinosus from B. bufo. This disparity, which gives further support to species separation, can probably be ascribed to changes in the allometric trajectories, and seen in light of the evolutionary history of the two lineages.
Keywords. Allometry, Bufo bufo, Bufo spinosus, divergence, geometric morphometrics, landmark, ontogeny, skull shape.
INTRODUCTION Scandinavia and eastwards deep into Russia; B. spinosus is found in the Iberian Peninsula, western and south- Divergent traits can be expected to be found in ern France, and North Africa, from Morocco to Tuni- closely related species with a broad geographical distri- sia (Arntzen et al., 2013a). Their lineages have diverged bution, relatively to ecological and climatic variation, around 9 Ma (million years ago; Recuero et al., 2012), but developmental constraints may also play an impor- but in contrast to a deep genetic differentiation, B. bufo tant role in restricting or channelling phenotypic evolu- and B. spinosus appear phenotypically similar (Arntzen tion (Cvijanović et al., 2014; Ivanović and Arntzen, 2018). et al., 2013a). Whereas a few diagnostic characters were The Common toad, Bufo bufo (Linnaeus, 1758), and described for the external morphology, virtually nothing the Spined toad, Bufo spinosus Daudin, 1803, are mem- is known about interspecific osteological differences, nei- bers of the Common toad species group of the western ther in the postcranial nor in the cranial skeleton. Such a Palaearctic (Arntzen et al., 2013b). B. bufo has a wide complex structure as the skull is of particular interest in Eurasian distribution that comprises northern and east- a wide range of studies, due to its fundamental biologi- ern France, central and southern Europe (including Sar- cal functions and the fact that it often undergoes adaptive dinia; Cossu et al., 2018), and stretches northwards into variation (Ivanović et al., 2012). Substantial changes are
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 118 Giovanni Sanna known to occur in the anuran skull after metamorphosis Landmarks (Ponssa and Candioti, 2012). The purpose of this investigation was to highlight Thirty-one homologous landmark points were collected on potential interspecific differences in the skull morphology each 3D surface model to describe overall skull shape (Fig. 1), using Landmark editor 3.6 (Institute for Data Analysis and Vis- of B. bufo and B. spinosus, in the context of postmeta- ualization, University of California, Davis, 2007). Fifteen points morphic development, using a geometric morphometrics were bilateral and symmetric, while one was median. Anatomi- approach. The analysis of ontogenetic shape variation in cal description of points is provided in Table S2. Incomplete these two species is interesting not only for a taxonomic skull ossification of small juveniles made it challenging, at evaluation based on morphology; it can provide insights times, to perform an accurate placement of homologous land- into their morphological evolution, to be interpreted in marks (Zelditch et al., 2004). the context of their evolutionary history and past distri- bution. Moreover, this analysis could contribute to a bet- ter understanding of the interplay between ontogeny and Geometric morphometrics morphological differentiation among anuran species. Morphometric and statistical analyses based on the land- mark coordinates, and qualitative observation of the associated shape changes, were carried out with MorphoJ 1.06 software MATERIAL AND METHODS (Klingenberg, 2011). A generalized Procrustes analysis (GPA), consisting of a full Procrustes superimposition for object sym- A total of 58 alcohol-stored specimens, including freshly metry, was applied: the symmetric components of shape varia- metamorphosed, juvenile, and adult (male and female) toads, tion, as a measure of skull shape, and centroid size, as a meas- were analysed (Table S1). These toads had been collected from ure of skull size, were computed for each individual (Klingen- various populations in different localities of the Iberian Penin- berg, 2016). The covariance matrix of shape variables was then sula (B. spinosus), France (B. spinosus, B. bufo), and the Neth- generated and used to perform a principal component analysis erlands (B. bufo). They were subdivided into two ontogenetic (PCA), in order to explore overall patterns of shape variation. series, made up of 28 Bufo bufo and 30 Bufo spinosus individu- Shape changes in relation to growth were evaluated with als, respectively. Body size in the whole sample ranged from a multivariate regression of shape (symmetric components, i.e. 16.0 mm to 78.0 mm SUL (snout-urostyle length). dependent variables) on size (log-transformed centroid size, i.e. Mature B. spinosus individuals were larger on average independent variable), one for each ontogenetic series (Mon- (mean SUL 71.5 mm) than B. bufo ones (mean SUL 60.4 mm), teiro, 1999), in association with a permutation test against inde- reflecting size disparity between western European Common pendence between dependent and independent variables (made toads and Spined toads (Cvetković et al., 2009). up of 10⁴ randomization rounds). The angle between the two regression vectors was calculated for comparison, including a test against randomness of vector directions in shape tangent Three-dimensional imaging space. Interspecific morphological distinction was assessed by Skulls were CT-scanned with two x-ray machines: SkyScan performing a discriminant function analysis (DFA) with leave- 1172 (Bruker, Kontich, Belgium) and ZEISS Xradia 520 Versa one-out cross-validation (Lachenbruch, 1967; Webster and (Carl Zeiss XRM, Pleasanton, CA, USA). The former was used Sheets, 2010) on shape data of the 25 largest individuals, 12 B. for the smaller toads, with 0.5 mm Aluminium filter, 2K reso- bufo and 13 B. spinosus toads comprising adults and subadults. lution (2000 × 1336), pixel size of 13.17 µm, voltage of 29-54 Such a subsample was selected in order to maximize interspe- kV, exposure time of 420-750 ms, 0.4 rotation step, averaging cific differences, since the PCA had previously shown shape of four frames. Voltage and exposure time were modified when divergence in late stages of growth (Fig. 2). Results of this anal- scanning specimens of different sizes (which showed differences ysis were compared with those of a DFA on the remaining indi- in skull density), in order to keep image quality consistent. Xra- viduals of the sample, namely juveniles. Both analyses included dia scanner was employed for adult individuals, mainly due to a parametric T-square test against the null hypothesis of equal its larger sample holder; resolution was set at 1K (1000 × 1024), group means, and a permutation test for the T-square statistic with pixel size of 34.18 µm, and voltage of 80 kV. with 10,000 randomization rounds. Scanning was followed by two-dimensional reconstruction of raw image data into stacks, which were processed with Avizo 9 software (FEI SAS, France): the segmentation editor was used RESULTS to segregate homogeneous volumes (corresponding to skull bones), with manual adjustment of masking. Notable variation in the extent of cranial ossification, not merely restricted to The first two principal components (PC1, PC2) of small juveniles, was observed at this point. A 3D surface model the PCA accounted for 71.9% of total shape variation. of the skull was then generated, applying a variable degree of PC1 alone grouped 65.4% of variation and was positive- unconstrained smoothing according to ossification extent. ly correlated with size: therefore, the scatter plot of PC2 Skull shape variation in Bufo toads 119
Fig. 2. Ontogenetic shape variation of B. bufo and B. spinosus, described by the scatter plot of the first two principal components (with 95% confidence ellipses). PC1 summarises allometric varia- tion, while PC2 displays species divergence in late development.
vs. PC1 can be looked at as a morphospace that contains the ontogenetic shape trajectories of B. bufo and B. spi- nosus (Fig. 2). Cranial shape variation along PC1 was characterised, in the positive direction, by a broadening (at the level of the jaw joint), an overall shortening (at the level of both the exoccipital and the premaxilla), and dorsoventral compression. Some interspecific variation was comprised in PC2, especially for the largest toads; positive direction shape changes along this axis were an increase in skull length, dorsoventral compression, and a slight widening. Both regression analyses found a significant asso- ciation between shape and size (P < 0.0001 for both), indicating allometric growth: 63.8% of shape variation in B. bufo and 68.5% of shape variation in B. spinosus were predicted by regression on size (Fig. 3). The two vector directions formed an angle of 17.2° (whereas 0° denote complete correspondence, 90° maximum diver- gence) and were not random in the shape tangent space (P < 0.00001). The major shape changes associated with regression were corresponding between the two species and matched those related to the first principal compo- nent of the PCA, thus confirming the correlation of PC1 with skull size. Discriminant function analysis applied to the largest toads showed a clear distinction between the two species (Fig. 4): after cross-validation, 10/12 B. bufo individuals were reassigned to their true group and 2/12 were allo- cated to the Spined toad group, while 13/13 B. spinosus individuals were reassigned to their own group (Cohen’s K = 0.84). However, the T-square test for the difference between group means was not statistically significant (T² = 281, P = 0.82). Permutation produced a signifi- Fig. 1. Landmark positions on the digitized skulls, in dorsal (A), right lateral (B), and frontal (C) views. For the anatomical descrip- cant result instead (P < 0.0001). The major interspecific tion of landmarks, see Table S2. shape differences pointed out by the analysis concerned cranial length and height: B. spinosus exhibited a longer upper jaw (with a more posterior jaw joint), yet a shorter 120 Giovanni Sanna
Fig. 3. Shape changes of B. bufo (A) and B. spinosus (B) in relation to size, described by the scatter plot of regression scores against log- transformed centroid size.
Fig. 4. Interspecific distinction, described by DFA histograms in which light grey bars represent B. bufo and dark grey bars represent B. spi- nosus. Species assignment and discriminant scores are shown before (A, C) and after (B, D) cross-validation, for the largest individuals (A, B) and juveniles (C, D). skull (due to a shorter occipital region), with a dorsally (i.e., the association between morphological changes and expanded snout; skull width did not show noteworthy ontogenetic growth; Klingenberg, 2016) characterises variation. DFA on juveniles yielded similar results, with a skull development of both Bufo bufo and Bufo spinosus, weaker interspecific distinction (Fig. 4): 13/16B. bufo and mainly in the form of a considerable widening (typi- 14/17 B. spinosus toads were reallocated to their group (K cal of skull growth in frogs; Ponssa and Candioti, 2012), = 0.63); T-square test was not significant (T² = 790.5, P = and a relative shortening and dorsoventral flattening. 0.72), while permutation test was significant (P < 0.0001). Cranial allometry has already been recognized in other anuran species, both in larval (e.g., in Rana sylvatica; Larson, 2002) and post-metamorphic development (e.g. DISCUSSION in Rhinella marina and the Leptodactylus fuscus group; Birch, 1999; Ponssa and Candioti, 2012). Allometric con- Most of the shape variation in the whole dataset was straints are likely to limit phenotypic evolution of the explained by differences in size, as shown by the con- skull, even in presence of selective pressure (Simon et al., cordant results of principal component and regression 2016). This kind of scenario can be logically applied to analyses. Therefore, I suggest that ontogenetic allometry the current study, which found shape disparities between Skull shape variation in Bufo toads 121
B. bufo and B. spinosus that are moderate if compared to availability (Simon et al., 2016), and changes in these eco- the common allometric variation. logical factors may lead to skull shape divergence – even Nevertheless, there seems to be a shift in the ontoge- in closely related species. Alternative climate-driven fac- netic trajectories of the two species, highlighted by both tors could also induce skull differentiation, such as repro- the principal component analysis (Fig. 2) and the angle duction sites (water pools) occurrence, and the ability of between regression vectors (17.2°), and reflected in the toads to detect them, which involves the olfactory cap- morphological distinction provided by the discrimi- sules in the snout region (Trueb, 1993; Simon et al., 2016). nant function analyses (Fig. 4). Along with both latter Whether such factors are related to the interspecific dif- analyses, T-square test against equal species mean shapes ferences found here – some of which concern the snout, yielded significant results only after permutation, which more expanded in B. spinosus – it is not possible to say. probably compensated for the relatively small sample Arntzen et al. (2013a) hypothesized that in European size. Interspecific divergence, however, resulted higher toads a larger body size, a more pronounced presence of in the group of largest individuals: they showed greater keratinous spines on the cheek warts and a wider head agreement to true species membership, as described by shape might favour defence against predators, namely a K coefficient of 0.84 against the 0.63 of juveniles. Con- grass snakes. Although B. spinosus apparently meets these sequently, it might be easier to discriminate between B. requirements more than B. bufo, relative cranial width did bufo and B. spinosus at mature stages, rather than among not show significant interspecific disparity in the current young individuals. This would be in contrast with the study. Thus, no link was found between widely divergent findings from other studies on amphibians, which point- parotoid glands, a distinctive trait of B. spinosus, and a ed out a decrease in interspecific disparity over ontogeny wider skull. Nevertheless, Spined toads could compensate (Adams and Nistri, 2010; Ponssa and Candioti, 2012). by attaining a wider skull through their bigger size. DFA and, to some extent, PCA, indicate that B. bufo For further assessments of drivers and extent of toads have on average a longer skull and a more dorsally morphological divergence between B. bufo and B. spino- compressed snout than B. spinosus toads. This morpho- sus, it would be convenient to use a larger sample, espe- logical divergence could theoretically be placed within cially for adult individuals, in order to properly account either the non-allometric or the allometric regime. Vari- for the role of sexual dimorphism and benefit from the ation in non-allometric shape would imply some sort of highest disparity. Different populations should be consid- relaxation of ontogenetic constraints, which seems very ered, since B. bufo exhibits remarkable variation across its unlikely, as the present evidence suggests that these con- wide range; for instance, Mediterranean Common toads straints are strong in both species. Alternatively, inter- appear to resemble Spined toads in body size – large preting interspecific variation as comprised in the allo- – and skin texture – thick and warty (De Lange, 1973; metric framework should better reflect the results of the Cvetković et al., 2009; Arntzen et al., 2013a). analyses. Allometric variation could have arisen in two ways. First, from changes in developmental timing of the ancestral shape trajectory, with a conserved shape- ACKNOWLEDGEMENTS size relationship (ontogenetic scaling hypothesis; Strelin et al., 2016); however, this also seems unlikely, because This study was conducted in association with Natura- the divergent skull shapes do not correspond to different lis Biodiversity Center. I would like to thank Marta Cal- stages of the same trajectory. Secondly – and more plau- vo (Museo Nacional de Ciencias Naturales) and Esther sibly – divergent evolution of the ontogenetic programme Dondorp (Naturalis Biodiversity Center) for access to in the two lineages may have occurred, with a conserved the material under their supervision, Ana Ivanović for direction of early post-metamorphic shape trajectories. precious suggestions on study design, Rob Langelaan for The latter hypothesis would imply that interspecific dif- technical support with computed tomography, and Pim ferences have arisen during the long-lasting separation of Arntzen and two anonymous reviewers for valuable feed- the lineages (about 9 Ma), and it could even be surprising back and remarks on earlier versions of the manuscript. that they are not more pronounced; as a possible expla- nation for this moderate divergence, a recent morpho- logical evolution, following rapid postglacial European SUPPLEMENTARY MATERIAL expansion of B. bufo from its Balkan refugium, cannot be excluded (Recuero et al., 2012; Arntzen et al., 2016). Supplementary material associated with this article Climate is thought to have an indirect influence on can be found at
REFERENCES Lachenbruch, P.A. (1967): An almost unbiased method of obtaining confidence intervals for the probability of Adams, D.C., Nistri, A. (2010): Ontogenetic convergence misclassification in discriminant analysis. Biometrics and evolution of foot morphology in European cave 23: 639-645. salamanders (Family: Plethodontidae). BMC Evol. Larson, P.M. (2002): Chondrocranial development in lar- Biol. 10: 216. val Rana sylvatica (Anura: Ranidae): morphometric Arntzen, J.W., McAtear, J., Recuero, E., Ziermann, J.M., analysis of cranial allometry and ontogenetic shape Ohler, A., van Alphen, J., Martínez-Solano, I. (2013a): change. J. Morphol. 252: 131-144. Morphological and genetic differentiation of Bufo Monteiro, L.R. (1999): Multivariate regression models toads: two cryptic species in Western Europe (Anura, and geometric morphometrics: the search for causal Bufonidae). Contrib. Zool. 82: 147-169. factors in the analysis of shape. Syst. Biol. 48: 192-199. Arntzen, J.W., Recuero, E., Canestrelli, D., Martínez- Ponssa, M.L., Candioti, F.V. (2012): Patterns of skull Solano, I. (2013b): How complex is the Bufo bufo spe- development in anurans: size and shape relationship cies group? Mol. Phylogenet. Evol. 69: 1203-1208. during postmetamorphic cranial ontogeny in five spe- Arntzen, J.W., Trujillo, T., Butôt, R., Vrieling, K., Schaap, cies of the Leptodactylus fuscus Group (Anura: Lepto- O., Gutiérrez-Rodríguez, J., Martínez-Solano, I. dactylidae). Zoomorphology 131: 349-362. (2016): Concordant morphological and molecular Recuero, E., Canestrelli, D., Vörös, J., Szabó, K., Poyark- clines in a contact zone of the Common and Spined ov, N.A., Arntzen, J.W., Crnobrnja-Isailovic, J., Kidov, toad (Bufo bufo and B. spinosus) in the northwest of A.A., Cogălniceanu, D., Caputo, F.P., Nascetti, G., France. Front. Zool. 13: 52. Martínez-Solano, I. (2012): Multilocus species tree Birch, J.M. (1999): Skull allometry in the marine toad, analyses resolve the radiation of the widespread Bufo Bufo marinus. J. Morphol. 241: 115-126. bufo species group (Anura, Bufonidae). Mol. Phylo- Cossu, I.M., Frau, S., Delfino, M., Chiodi, A., Corti, C., genet. Evol. 62: 71-86. Bellati, A. (2018): First report of Bufo bufo (Linnaeus, Simon, M.N., Machado, F.A., Marroig, G. (2016): High 1758) from Sardinia (Italy). Acta Herpetol. 13: 43-49. evolutionary constraints limited adaptive responses Cvetković, D., Tomašević, N., Ficetola, G.F., Crnobrnja- to past climate changes in toad skulls. Proc. R. Soc. B Isailović, J., Miaud, C. (2009): Bergmann’s rule in Biol. Sci. 283: 20161783. amphibians: combining demographic and ecological Strelin, M.M., Benitez-Vieyra, S., Fornoni, J., Klin- parameters to explain body size variation among pop- genberg, C.P., Cocucci, A.A. (2016): Exploring the ulations in the common toad Bufo bufo. J. Zool. Syst. ontogenetic scaling hypothesis during the diversifica- Evol. Res. 47: 171-180. tion of pollination syndromes in Caiophora (Loasace- Cvijanović, M., Ivanović, A., Kalezić, M.L., Zelditch, M.L. ae, subfam. Loasoideae). Ann. Bot. 117: 937-947. (2014): The ontogenetic origins of skull shape dispar- Trueb, L. (1993): Patterns of cranial diversity among the ity in the Triturus cristatus group. Evol. Dev. 16: 306- Lissamphibia. In: The skull: patterns of structural and 317. systematic diversity, pp. 255-343. Hanken, J., Hall, De Lange, L. (1973): A contribution to the intraspecific B.K., Eds, University of Chicago Press, Chicago. systematics of Bufo bufo (Linnaeus, 1758) (Amphibia). Webster, M., Sheets, H.D. (2010): A practical introduc- Beaufortia 21: 99-116. tion to landmark-based geometric morphometrics. Ivanović, A., Arntzen, J.W. (2018): Evolution of skull Paleontol. Soc. Pap. 16: 163-188. shape in the family Salamandridae (Amphibia: Cau- Zelditch, M.L., Swiderski, D.L., Sheets, H.D., Fink, W.L. data). J. Anat. 232: 359-370. (2004): Geometric morphometrics for biologists: a Ivanović, A., Sotiropoulos, K., Üzüm, N., Džukić, G., primer. Elsevier/Academic Press, Amsterdam. Olgun, K., Cogălniceanu, D., Kalezić, M.L. (2012): A phylogenetic view on skull size and shape variation in the smooth newt (Lissotriton vulgaris, Caudata, Sala- mandridae). J. Zool. Syst. Evol. Res. 50: 116-124. Klingenberg, C.P. (2011): MorphoJ: an integrated soft- ware package for geometric morphometrics. Mol. Ecol. Resour. 11: 353-357. Klingenberg, C.P. (2016): Size, shape, and form: concepts of allometry in geometric morphometrics. Dev. Genes Evol. 226: 113-137. Acta Herpetologica 14(2): 123-128, 2019 DOI: 10.13128/a_h-7750
Hematological parameters of the Bolson tortoise Gopherus flavomarginatus in Mexico
Cristina García-De la Peña1,*, Roger Iván Rodríguez-Vivas2, Jorge A. Zegbe-Domínguez3, Luis Manuel Valenzuela-Núñez1, César A. Meza Herrera4, Quetzaly Siller-Rodríguez1, Verónica Ávila-Rodríguez1 1 Facultad de Ciencias Biológicas, Universidad Juárez del Estado de Durango, Gómez Palacio, Durango, México. C.P. 35010. *Corre- sponding author. Email: [email protected] 2 Facultad de Medicina Veterinaria y Zootecnia, Universidad Autónoma de Yucatán, Mérida, Yucatán, México, C.P. 97100 3 Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias, Campo Experimental Calera, Zacatecas, México, C.P. 98500 4 Universidad Autónoma Chapingo-URUZA, Bermejillo, Durango, México, C.P. 35230
Submitted on: 2018, 21th June; Revised on: 2019, 10th May; Accepted on: 2019, 23rd August Editor: Emilio Sperone
Abstract. We present findings of our preliminary study to determine biometry and blood chemistry values of healthy wild individuals of the critically endangered Bolson tortoises (Gopherus flavomarginatus) in Mexico. Given the absence of previously published data regarding hematology parameters for this species, these results represent an important base for additional research. Hematocrit determination, stains, and cell counts were performed, as well as 18 parameters of blood chemistry. Values of biometry and blood chemistry for G. flavomarginatus were similar to ref- erence values those already reported for G. agassizii, G. polyphemus, and G. berlandieri. These similarities reflect the phylogenetic relationships among these species. However, slight differences may point to particular adaptations that each has developed to their own habitat, and so point to questions to be addressed with future research.
Keywords. Chelonia, lymphocyte, desert, Mapimí Biosphere Reserve.
Hematological analysis is a common technique for currently has protected status within the Mapimí Bio- evaluating the health status of wild animals (Campbell, sphere Reserve (CONANP, 2006). Tortoises of the genus 2004; Tavares-Dias et al., 2009). Blood biometry and Gopherus are keystone organisms for the ecosystems in chemistry of terrestrial, semiaquatic and marine tortoises which they live because, due to their feeding habits (her- have been described already (Stacy et al. 2011; Camp- bivory), they perform the ecological function of seed bell, 2015). However, there are still species of ecologi- dispersal (Carlson et al., 2003), and because the burrows cal importance whose blood information is unavailable. that they excavate are deep and provide shelter for at least This is the case of Gopherus flavomarginatus, a tortoise 300 species of invertebrates and 60 of vertebrates (Lips, endemic to Mexico and considered the largest terrestrial 1991). Due to the importance of protecting this tortoise, tortoise in North America, with a carapace length of up we conducted this preliminary study to determine biom- to 40 cm (Morafka et al., 1989). This species is in dan- etry and blood chemistry values of healthy wild indi- ger of extinction according to the Official Mexican Stand- viduals. This information will serve as a basis for future ard 059 (SEMARNAT, 2010) and critically endangered hematological studies carried out on this tortoise. according to the IUCN red list (Kiester et al., 2018). Its From May 2015 to September 2017, we captured 44 geographical distribution is restricted to the Mapimí adult individuals of G. flavomarginatus (16 males and 28 Bolson in the Mexican Chihuahuan Desert, where it females) within the Mapimí Biosphere Reserve in Mex-
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 124 Cristina García-De la Peña et alii ico (26°00’ and 26°10’N, and 104°10’ and 103°20’W). The H:L ratio was obtained by dividing the per- Blood samples were collected from the subvertebral centage of heterophiles by the percentage of lympho- vein. Three milliliters of blood were collected from each cytes in each sample (Davis et al., 2008). Blood from specimen; one milliliter was placed in a Vacutainer® tube the red Vacutainer® tube was centrifuged at 12,000 g with lithium heparin as anticoagulant and the rest in a for five minutes; the serum was separated from the red red Vacutainer® tube. The tubes were stored in a cooler cells and placed in a sterile plastic tube. No hemoly- at a temperature of approximately 4 °C. Each tortoise sis was observed in any sample. The following param- was determined to be healthy following the observa- eters were analyzed: glucose, uric acid, urea, blood urea tion protocols of Jacobson (2014) and USFWS (2016). nitrogen (BUN), creatinine, total proteins, albumin, Tortoises were then released at the site of their capture. globulins, cholesterol, triglycerides, alanine aminotrans- Biometric analysis of blood was performed following the ferase (ALT), aspartate aminotransferase (AST), alkaline protocols suggested by Thrall et al. (2006) and Turgeon phosphatase (AP), chlorine, sodium, calcium, phospho- (2012). The volume percentage of red cells (Hematocrit, rus and osmolality (mOsm/kg=1.86 (Na [mmol/L]) + Ht in%) was determined in the blood contained in the glucose [mg/dL]/18 + BUN [mg/dL]/2.8 + 9). Human green Vacutainer® tube, using the microhematocrit tech- serum Spintrol H Spinreact® was used as control. These nique. The total number (TR) and percentage (PR) of red analyzes were performed in a VetTest® spectrophotome- cells were obtained for each blood sample. The formula ter with Spinreact® reagents. Descriptive statistics (mean, used to obtain hematocrit values was Ht% = (PR/TR) × standard error, median, standard deviation, minimum, 100. Hemoglobin concentration was quantified with the and maximum) were obtained in PAST 3.14 (Hammer et Drabkin colorimetric method, using the Spinreact® com- al., 2001). mercial kit, in a VetTest® spectrophotometer. The reaction Blood biometry results for G. flavomarginatus are product was centrifuged (12,000 g × 5 min) to precipitate shown in Table 1 and for blood chemistry in Table 2. A the nucleus of the cells and keep them from distorting comparison among biometry and chemistry variables of the color to be measured (Thrall et al., 2006). G. flavomarginatus and other Gopherus species are shown Erythrocyte and leukocyte counts were carried out in Tables 3 and 4, respectively. Blood biometric and using a Thoma pipette with red bead and a Neubauer chemistry values obtained for G. flavomarginatus were hemocytometer with Natt and Herrick’s stain (Thrall et al., similar to those reported for G. agassizii (Christopher et 2006). The erythrocyte count was carried out under 10× al., 1999; Dickinson et al., 2000), G. polyphemus (Taylor magnification, on both sides of the Neubauer chamber, in and Jacobson, 1982), and G. berlandieri (Teare, 2013a) the four corner squares and the center square within the (Tables 3 and 4). Christopher et al. (1999) reported that large center square of the chamber. The following formula the most abundant leukocytes in G. agassizii are het- was applied (Kemal, 2014): erythrocytes (×106 ul) = mean erophils, followed by lymphocytes, basophils, and finally erythrocytes × 10000. The leukocyte count was carried eosinophils and monocytes. In our findings for G. fla- out under 40x magnification in the nine large squares on vomarginatus, the order of abundance of the leukocyte both sides of the Neubauer chamber. The following for- cells was similar to the one mentioned by Duguy (1970), mula was applied: leukocytes (×103 ul) = mean leukocytes with lymphocytes as the most abundant leukocytes × 50 (Kemal, 2014). The erythrocyte indices were calcu- (males = 53.43%, 4.79 × 103 cel/ul; females = 44.03%, lated according to the formulas described by Ball (2014) 4.20 × 103 cel/ul). This agrees with the findings of Diaz- and Kemal (2014): mean corpuscular volume, MCV (fL) Figueroa (2005) for G. polyphemus, and other authors = Ht% × 10/erythrocytes (×106 ul); mean corpuscular who indicate that lymphocytes are predominant in the hemoglobin, MCH (pg) = (Hg × 10)/erythrocytes (×106 peripheral blood of most reptile species (Davis et al.2008; ul); mean corpuscular hemoglobin concentration, MCHC Stacy et al., 2011). (g/dl) = (Hg × 100)/Ht%. Two smears were prepared from Heterophils were the second most abundant leu- each sample on glass slides using Wright’s dye (Analyty- kocyte cells in G. flavomarginatus. In general, the abun- ka®). One hundred leukocytes were counted in the body dance of this type of leukocyte in reptiles ranges from 30 of the smear of each of the two slides. The average of both to 45% (Duguy, 1970; Frye, 1991); however, in tortoises slides was obtained, and the results were expressed as the it can reach up to 50% (Alleman et al., 1992; Christo- proportion of each cell type (relative differential count). pher et al. 1999). Basophils were the third most abundant To obtain the absolute differential count, the total value leukocytes in the blood of the Bolson tortoise. Alleman of leukocytes (×103 cel/ul) was multiplied by the average et al., (1992), and Duguy, (1970) estimated that the per- percentage of each type of leukocyte and the result was centage of basophils in healthy terrestrial and aquatic divided by 100 (Thrall et al., 2006). tortoises can be higher than 40%. In G. flavomargi- Hematological parameters of Gopherus flavomarginatus 125
Table 1. Descriptive statistics of blood biometric variables from male/female Gopherus flavomarginatus. SE = standard error, SD = standard deviation, Min = minimum, Max = maximum.
Variable Mean SE Median SD Min Max Hematocit (%) 25.35/23.48 1.41/0.77 25.30/23.61 5.64/4.09 16.47/17.89 35.62/33.33 Hemoglobin (g/dl) 6.53/6.59 0.40/0.26 6.56/7.15 1.62/1.41 4.05/3.89 9.67/8.95 Erythrocytes (×106 cel/ul) 0.54/0.57 0.06/0.03 0.47/0.58 0.25/0.19 0.22/0.32 0.97/0.98 MCV (fl) 539.02/448.98 54.81/27.17 458.56/401.53 219.26/143.78 264.56/196.17 921.72/717.91 MCH (pg) 144.12/122.92 18.96/5.79 124.11/122.53 75.86/30.67 64.66/45.95 292.86/197.78 MCHC (g/dl) 26.37/28.49 1.64/1.23 25.62/28.84 6.56/6.52 17.69/16.20 39.36/44.66 Leukocytes (×103 cel/ul) 8.88/9.50 0.82/0.70 8.55/8.55 3.28/3.73 5.00/2.44 17.67/21.33 Heterophils (%) 25.43/29.32 1.99/1.68 26.00/29.00 7.99/8.89 14.00/13.00 39.00/45.00 Eosinophils (%) 2.75/3.17 0.57/0.51 2.00/2.50 2.29/2.74 0.00/0.00 7.00/11.00 Basophils (%) 17.68/21.42 2.11/1.67 15.50/18.50 8.47/8.86 6.00/8.00 29.00/46.00 Lymphocytes (%) 53.43/44.03 2.99/2.06 54.00/42.50 11.99/10.92 34.00/27.00 73.00/70.00 Monocytes (%) 0.68/2.03 0.17/0.40 1.00/1.00 0.70/2.16 0.00/0.00 2.00/9.00 H:L ratio 0.52/0.73 0.06/0.06 0.45/0.71 0.27/0.36 0.22/0.24 1.11/1.59 Heterophils (×103 cel/ul) 2.17/2.81 0.19/0.29 2.09/2.56 0.77/1.55 0.93/0.66 3.45/7.25 Eosinophils (×103 cel/ul) 0.21/0.34 0.04/0.08 0.18/0.21 0.17/0.45 0.00/0.00 0.66/2.35 Basophils (×103 cel/ul) 1.61/1.95 0.28/0.16 1.45/1.80 1.14/0.85 0.30/0.39 5.12/4.16 Lymphocytes (×103 cel/ul) 4.79/4.20 0.57/0.36 3.96/4.04 2.31/1.95 2.57/0.78 10.32/8.96 Monocytes (×103 cel/ul) 0.07/0.18 0.02/0.03 0.06/0.09 0.09/0.20 0.00/0.00 0.35/0.85
Table 2. Descriptive statistics of blood chemistry variables from male/female Gopherus flavomarginatus. SE = standard error, SD = standard deviation, Min = minimum, Max = maximum.
Variable Mean SE Median SD Min Max Glucose (mg/dl) 79.19/58.34 7.97/8.28 68.33/40.47 31.90/43.82 43.32/13.34 134.67/165.30 Uric acid (mg/dl) 5.63/5.58 1.07/0.71 5.66/5.96 4.28/3.76 0.27/0.24 12.45/13.48 Urea (mg/dl) 25.60/31.22 0.82/0.76 25.39/31.75 3.30/7.86 20.64/14.67 32.86/51.34 BUN 11.96/14.37 0.38/0.82 11.86/14.22 1.54/4.35 9.64/6.86 15.36/23.99 Creatinine (mg/dl) 0.48/0.43 0.06/0.05 0.47/0.40 0.24/0.26 0.01/0.01 0.95/0.89 Total protein (g/dl) 4.12/3.95 0.53/0.24 3.76/3.78 2.13/1.27 1.29/2.00 8.96/7.13 Albumin (g/dl) 0.87/0.68 0.18/0.10 0.56/0.47 0.74/0.56 0.09/0.04 2.50/2.55 Globulins (g/dl) 3.24/3.27 0.42/0.24 3.23/3.09 1.70/1.30 0.99/1.37 6.46/6.50 Cholesterol (mg/dl) 280.55/214.94 45.28/23.50 269.50/167.43 181.13/124.36 41.46/78.32 692.27/508.20 Triglycerides (mg/dl) 207.85/177.57 13.79/16.18 213.80/164.06 55.18/85.62 114.08/42.35 283.63/402.30 ALT (UI/I) 7.74/8.15 1.06/1.06 7.53/6.48 4.25/5.65 2.04/1.74 15.93/22.43 AST (UI/I) 41.79/45.23 6.30/3.53 32.27/47.92 25.21/18.69 14.95/15.63 105.60/80.42 AP (UI/I) 60.30/66.53 8.08/4.58 56.91/70.05 32.32/24.24 14.11/24.86 102.60/99.32 Chlorine (mmol/l) 119.90/121.64 4.24/2.98 118.90/122.31 16.97/15.81 95.68/94.53 139.62/149.14 Sodium (mmol/l) 137.53/138.24 3.20/2.47 136.24/136.42 12.80/13.11 112.36/117.90 160.90/177.60 Calcium (mg/dl) 11.61/13.44 0.65/0.41 12.64/13.29 2.63/2.19 7.46/10.27 15.67/18.30 Phosphorus (mg/dl) 3.95/4.08 0.56/0.36 3.59/4.29 2.24/1.93 1.02/1.31 7.18/7.16 Osmolality (mOsm/kg) 270.50/273.87 5.92/4.73 268.60/272.56 23.70/23.85 225.43/240.81 316.39/345.66
natus, the average was 17.68% for males and 21.42% for of diet can influence the amount of eosinophilic leuko- females, with wide variation between individuals (min cytes in some species (Duguy, 1970; Deem et al., 2006). = 6%, max = 46%). Eosinophilic leukocytes were the According to Frye (1991), the percentage of eosinophils fourth most abundant leukocytes. The seasons and type in healthy reptiles ranges from 7 to 20%; in the case of 126 Cristina García-De la Peña et alii
Table 3. Comparison among minimum-maximum blood biometry values reported for Gopherus flavomarginatus and reference values of other three species of the genus.
Variable G. flavomarginatus1 G.agassizii2 G. polyphemus3,4 G. berlandieri5 Hematocit (%) 16.47-35.62 19.50-37.10 15.00-30.00 12.00-44.80 Hemoglobin (g/dl) 3.89-9.67 4.10-9.90 4.20-8.60 - Erythrocytes (×106 cel/ul) 0.22-0.98 0.36-1.08 0.24-0.91 - MCV (fl) 196.17-921.72 254.00-638.00 200.10-838.60 - MCH (pg) 45.95-292.86 74.00-186.00 - - MCHC (g/dl) 16.20-44.66 20.00-33.00 - - Leukocytes (×103 cel/ul) 2.44-21.33 1.49-10.92 10.00-22.00 0.00-80.650 Heterophils (%) 13.00-45.00 - 10.00-57.00 - Eosinophils (%) 0.00-11.00 - - - Basophils (%) 6.00-46.00 - 2.00-11.00 - Lymphocytes (%) 27.00-73.00 - 32.00-79.00 - Monocytes (%) 0.00-9.00 - 3.00-13.00 - H:L ratio males 0.22-1.59 - - - Heterophils (×103 cel/ul) 0.66-7.25 0.71-7.15 0.00-6.59 0.00-5.21 Eosinophils (×103 cel/ul) 0.00-2.35 0.00-0.95 - - Basophils (×103 cel/ul) 0.30-5.12 0.06-3.57 0.02-0.92 0.00-1.31 Lymphocytes (×103 cel/ul) 0.78-10.32 0.63-2.74 0-4.15 0-3.12 Monocytes (×103 cel/ul) 0.00-0.85 0.00-0.32 - 0.00-0.44
1 Present study. 2 Christopher et al. (1999) in summer season. 3 Taylor and Jacobson (1982). 4 Teare (2013b). 5 Teare (2013a).
Table 4. Comparison among minimum-maximum blood chemistry values reported for Gopherus flavomarginatus and reference values of other three species of the genus.
Variable G. flavomarginatus1 G.agassizii2 G. polyphemus3,4 G. berlandieri5 Glucose (mg/dl) 13.34-165.30 65.00-186.00 55.00-128.00 9.00-157.00 Uric acid (mg/dl) 0.24-13.48 1.70-9.20 0.90-8.50 0.00-8.60 Urea (mg/dl) 14.67-51.34 - 1.00-130.00 - BUN 6.86-23.99 1.00-37.00 2.00-29.00 0.00-13.00 Creatinine (mg/dl) 0.01-0.95 0.20-0.40 0.10-0.40 - Total protein (g/dl) 1.29-8.96 2.30-5.30 1.30-4.60 1.20-7.70 Albumin (g/dl) 0.04-2.55 0.80-1.90 0.50-2.60 0.50-2.70 Globulins (g/dl) 0.99-6.50 1.30-3.90 0.50-4.60 0.60-5.10 Cholesterol (mg/dl) 41.46-692.27 33.00-381.00 19.00-150.00 Triglycerides (mg/dl) 42.35-402.30 7.00-603.00 - - ALT (UI/I) 1.74-22.43 1.00-5.00 2.00-57.00 - AST (UI/I) 14.95-105.60 15.00-123.00 57.00-392.00 0.00-265.00 AP (UI/I) 14.11-102.60 25.00-114.00 11.00-71.00 - Chlorine (mmol/l) 94.53-149.14 101.00-138.00 35.00-128.00 89.00-122.00 Sodium (mmol/l) 112.36-177.60 127.00-176.00 127.00-148.00 123.00-153.00 Calcium (mg/dl) 7.46-18.30 8.60-23.90 10.00-14.00 5.00-17.40 Phosphorus (mg/dl) 1.02-7.18 1.10-6.50 1.00-3.10 0.70-4.90 Osmolality (mOsm/kg) 225.43-345.66 252.00-352.00 - -
1Present study. 2Christopher et al. (1999) in summer season. 3Taylor and Jacobson (1982). 4Teare (2013b). 5Teare (2013a). Hematological parameters of Gopherus flavomarginatus 127
G. flavomarginatus, the average percentage found in the funding this study. To Magdalena Rivas-García for her present study was 2.75% for males and 3.17% for females, help in the field work. Cristino Villarreal-Wislar and the with a total variability of 0-11%. The average percentage Mapimí Biosphere Reserve personnel for logistical sup- of monocytes in the Bolson tortoise was 0.68% for males port during the realization of this study. To Cameron and 2.03% for females (min = 0, max = 9); these values W. Barrows (University of California, Riverside) for the are within the range of abundance reported for reptiles review of this manuscript. Tortoise samples were collect- (Duguy, 1970). ed under the DGVS 07249/15-16-17 permit granted by The H:L ratio (heterophils/leukocytes) has been used SEMARNAT, Mexico. as a reliable method to evaluate the exposure of verte- brates to chronic stress due to the relationship between the leukocyte profile and the adrenal response (produc- REFERENCES tion of cortisol). When the level of cortisol rises, the number of circulating heterophils increases, while the Alleman, A.R., Jacobs, E.R., Raskin, R.E. (1992): Morpho- number of lymphocytes decreases (Davis et al., 2011). logic and cytochemical characteristics of blood cells Davis (2009) indicated that blood biometry studies from the desert tortoise (Gopherus agassizii). Am. J. conducted with terrestrial tortoises in which the pos- Vet. Res. 53: 1645-1651. sibility of individuals being stressed was not considered Campbell, T.W. (2004): Hematology of lower vertebrates. (including studies with G. agassizii, G. berlandieri, and In: 55th Annual Meeting of the American College of G. polyphemus), the H:L ratio was less than 2.0 (mean of Veterinary Pathologists (ACVP) & 39th Annual Meet- 0.65, SD=0.34), which coincides with the values obtained ing of the American Society of Clinical Pathology, pp. in the present study for G. flavomarginatus. 1103-1214. ASVCP, ACVP, ASVCP, Eds, Middleton Blood chemistry variables provide important infor- WI, USA. mation for establishing levels of hydration (BUN, uric Campbell, T.W. (2015): Exotic Animal Hematology and acid, osmolality), nutrition (glucose, total protein, albu- Cytology. 4th ed. Wiley Blackwell, Oxford, UK. min, colesterol, phosphorus), and metabolic activity (ala- Carlson, J.E., Menges, E.S., Marks, P.L. (2003): Seed dis- nine aminotransferas, aspartate aminotransferase, and persal by Gopherus polyphemus at Archbold Biological alkaline phosphatase activities) of desert tortoises (Chris- Station, Florida. Fla. Sci. 66: 147-154. topher et al., 1999). Christopher et al. (1999) showed that Christopher, M.M., Berry, K.H., Wallis, I.R., Agy, K.A., blood chemistry values in G. agassizii change due to the Enen, B.T., Peterson, C.C., (1999): Reference intervals effect of physiological (reproductive cycle, hibernation) and physiologic alterations in hematologic and bio- and environmental factors (precipitation patterns, avail- chemical values of free-ranging desert tortoises in the ability of water and food). However, we observed similar Mojave Desert. J. Wildl. Dis. 35: 212-238. blood values between G. agassizii and G. flavomarginatus CONANP. (2006): Programa de Conservación y Manejo at the same times of year (summer). In general, the phylo- Reserva de la Biosfera Mapimí México. CONANP- genetic relation among G. flavomarginatus, G. agassizii, G. SEMARNAT, Mexico. polyphemus, and G. berlandieri (Reynoso and Montellano- Davis, A.K. (2009): The wildlife leukocytes webpage; the Ballesteros, 2004) may be reflecting the similarity between ecologists’ source for information about leukocytes of reported hematology values for these species, and the wildlife species. Available at: http://wildlifehematol- slight differences may point to particular adaptation con- ogy.uga.edu. August 17, 2017. ditions that each one has developed in their own habitat. Davis, A.K., Maney, D.L., Maerz J.C. (2008): The use of Our small sample size (n = 44) precluded the cal- leukocyte profiles to measure stress in vertebrates: a culation of formal reference intervals, a process that review for ecologists. Funct. Ecol. 22: 760-772. would require a minimum of 120 individuals (Geffre, et Davis, A.K., Ruyle, L.E., Maerz, J.C. (2011): Effect of trap- al., 2009). However, given the limited sample size avail- ping method on leukocyte profiles of Black-Chested able, and the lack of previously published data regard- Spiny-Tailed Iguanas (Ctenosaura melanosterna): ing hematology parameters in G. flavomarginatus, these implications for zoologists in the field. ISRN Zool. results provide a useful starting point for researchers. 2011: 1-8. Deem, S.L., Dierenfeld, E.S., Sounguet, G.P., Alleman, A.R., Cray, C., Poppenga, R.H., Norton, T.M., Karesh, ACKNOWLEDGMENTS W.B. (2006): Blood values in free-ranging nesting leatherback sea turtles (Dermochelys coriacea) on the To Fondo Sectorial de Investigación para la Edu- coast of the Republic of Gabon. J. Zoo. Wildl. Med. cación SEP-CONACYT Ciencia Básica (220658) for 37: 464-471. 128 Cristina García-De la Peña et alii
Dickinson, V.M., Jarchow, J.L., Trueblood, M.H. (2002): Reynoso, V.H., Montellano-Ballesteros, M. (2004): A new Hematology and plasma biochemistry reference range giant turtle of the genus Gopherus (Chelonia: Testu- values for free-ranging desert tortoises in Arizona. J. dinidae) from the Pleistocene of Tamaulipas, México, Wildl. Dis. 38: 143-153. and a review of the phylogeny and biogeography of Diaz-Figueroa, O. (2005): Characterizing the health status gopher tortoises, J. Vert. Paleontol. 24: 822-837. of the Louisiana Gopher tortoise (Gopherus polyphe- SEMARNAT. (2010): NORMA Oficial Mexicana NOM- mus). Unpublished doctoral dissertation. Louisiana 059-SEMARNAT-2010, Protección ambiental-Especies State University and Agricultural and Mechanical nativas de México de flora y fauna silvestres-Cate- College. gorías de riesgo y especificaciones para su inclusión, Duguy, R. 1970. Numbers of blood cells and their varia- exclusión o cambio-Lista de especies en riesgo. Diario tion. In: Biology of Reptilia (vol 3), pp. 93-109. Gans, Oficial, 30 Diciembre 2010, México. C., Pough, F.H., Eds, Academic Press, New York. Stacy, N.I., Alleman, A.R., Sayler, K.A. (2011): Diagnostic Frye, F.L. (1991): Infectious diseases. In: Biomedical and hematology of reptiles. Clin. Lab. Med. 31: 87-108. surgical aspects of captive reptile husbandry, pp. 113- Tavares-Dias, M., Oliveira-Junior, A.A., Silva, M.G., Mar- 123. Frye, F.L., Ed, Krieger Publishing Co., Malabar, con, J.L., Barcellos J.F.M. (2009): Comparative hema- Florida. tological and biochemical analysis of giant turtles Geffré, A., Friedrichs, K., Harr, K., Concordet, D., from the Amazon farmed in poor and normal nutri- Trumel, C., Braun, J.P. (2009): Reference values: a tional conditions. Vet. Arh. 79: 601-610. review. Vet. Clin. Path. 38: 288-298. Taylor, R.W., Jacobson, E.R. (1982): Hematology and Hammer, Ř., Harper, D.A.T., Ryan, P.D. (2001): PAST: serum chemistry of the gopher tortoise, Gopherus Paleontological statistics software package for educa- polyphemus. Comp. Biochem. Physiol. 72: 425-428. tion and data analysis. Palaeontol. Electron. 4: 9. Teare, J.A. (2013a): Gopherus berlandieri, no selection by Jackson, D., Milstrey, E.G. (1989): The fauna of gopher gender, all ages combined, Conventional American tortoise burrows. In: Proceedings of the Gopher Tor- units 2013 CD.html. In: ISIS Physiological Reference toise Relocation Symposium, pp. 86-98. Diemer, J., Intervals for Captive Wildlife: A CD-ROM Resource. Jackson, D., Landers, L., Layne, J., Wood, D., Eds, International Species Information System, Ed, Eagan, Florida Game and Freshwater Fish Comm. Nongame MN, USA. Wildlife Program, Technical Report No. 5. Tallahas- Teare, J.A. (2013b): Gopherus polyphemus, no selection see, Florida. by gender, all ages combined, Conventional American Jacobson, E.R. (2014): Health issues of North Ameri- units 2013 CD.html. In: ISIS Physiological Reference can tortoises. In: Biology and conservation of North Intervals for Captive Wildlife: A CD-ROM Resource. American tortoises, pp. 60-76. Rostal, D.C., McCoy, International Species Information System, Ed, Eagan, E.D., Mushinsky, H.R., Eds, Johns Hopkins University MN, USA. Press. Baltimore, Maryland. Thrall, M.A., Baker, D.C., Campbell, T.W., DeNicola, Kemal, J. (2014): Laboratory Manual and Review on D., Fettman, M.J., Lassen, E.D., Rebar, A., Weiser, G. Clinical Pathology. OMICS Group eBooks, California, (2006): Veterinary hematology and clinical chemistry. EUA. Blackwell Publishing, Iowa, USA. Kiester, A.R., Palomo-Ramos. R., Ríos-Arana, J., Goode, Turgeon, M.L. (2012): Clinical Hematology: Theory & E.V. (2018). Gopherus flavomarginatus. The IUCN Procedures. 5th ed. Wolters Kluwer, Lippincott Wil- Red List of Threatened Species; e.T9402A112660985. liams & Wilkins, Philadelphia, USA. Available at: http://dx.doi.org/10.2305/IUCN. USFWS (2016): Health assessment procedures for the UK.2018-2.RLTS.T9402A112660985.en. [accessed on desert tortoise (Gopherus agassizii): A handbook per- 3 December 2018] tinent to translocation. U.S. Fish and Wildlife Service, Lips, K.R. (1991): Vertebrates associated with tortoise Desert Tortoise Recovery Office, Reno, Nevada, USA. (Gopherus polyphemus) burrows in four habitats in south-central Florida. J. Herpetol. 25: 477-481. Morafka, D.J., Aguirre, G., Adest, G.A. (1989): Gopherus flavomarginatus Bolson Tortoise. In: The Conserva- tion Biology of Tortoises, pp. 10-13. Swingland, R., Klemens, M.W., Eds, Occasional Papers of the IUCN Species survival Commission (SSC) No. 5. IUCN, Gland, Switzerland. Acta Herpetologica 14(2): 129-133, 2019 DOI: 10.13128/a_h-7751
Visible Implant Alphanumeric (VIA) as a marking method in the lesser snouted treefrog Scinax nasicus
Andrea Caballero-Gini1,2,3,*, Diego Bueno Villafañe2,3, Lía Romero2, Marcela Ferreira2,3, Lucas Cañete4, Rafaela Laino2, Karim Musalem2,5 1 Insituto de Biología Subtropical, Universidad Nacional de Misiones, Félix de Azara 1552, CP 3300, Posadas, Misiones, Argentina. *Corresponding author. Email: [email protected] 2 Centro de Investigación del Chaco Americano, Fundación Manuel Gondra, San José 365, Asunción, Paraguay 3 Instituto de Investigación Biológica del Paraguay, Del Escudo, CP, Asunción, Paraguay 4 Universidad Nacional de Asunción, Facultad de Ciencias Exactas y Naturales (FACEN), Campus Universitario, San Lorenzo, Paraguay 5 World Wildlife Fund, Bernardino Caballero 191, Asunción, Paraguay
Submitted on: 2019, February 22nd; Revised on: 2019, June 3th; Accepted on: 2019, June 3th Editor: Daniele Pellitteri-Rosa
Abstract. In this study we assessed the efficacy of Visible Implant Alphanumeric (VIA) for marking adults and juve- niles of the Neotropical treefrog Scinax nasicus. We evaluated the success of this technique in the identification of individuals and the prevalence of tags in the field. As a control, we marked the same individuals through toe-clipping. Of 196 marked individuals, 57 were recaptured in a 7-month study period. Only one mark was unreadable because it was located too deep in the skin. We found one case of tag expulsion and two inverted tags. Almost complete regen- eration of the adhesive disk was observed by the fifth month of the study in all recaptured frogs. We suggest VIA tagging method as suitable for S. nasicus over long term studies. Even though, a hybrid method for marking (VIA + toe-clipping) is recommended for species with dark and/or loose skin, or large frogs.
Keywords. Amphibian, mark-recapture, fluorescent elastomer, toe-clipping.
The recognition of individuals in a population is a the last two manufactured by North-west Marine Tech- key aspect in many amphibian ecology and conserva- nology, Shaw Island, USA (Heard et al., 2008; Branelly et tion studies (Seber, 1982; Campbell et al., 2009; Clemas al., 2014). et al., 2009). Their identification over time and space Toe-clipping is the most widely used technique for helps researchers to estimate individual and demographic marking anurans and salamanders (Ferner, 2010). How- parameters such as growth and mortality rates, dispersal, ever, this method is highly criticized because a decrease population size and habitat use (Osbourne et al., 2011). in the probability of recapture has been detected in some In cases where individuals of a species do not possess amphibian species (Clarke, 1972; Mccarthy and Par- characteristics that distinguish them from each other, like ris, 2004; Waddle et al., 2008). Another disadvantage of coloration and color patterns, and photo-identification toe-clipping in long term studies is the possibility of tis- is not suitable, it is necessary to use marking techniques sue regeneration, although this has been observed only (Donelly et al., 1994). For amphibians, the most com- in a few species (Ferner, 2007; Ursprung et al., 2011). monly used are toe-clipping, Passive Integrated Tran- PIT tags consist of an electromagnetic capsule with an sponder (PIT), Visible Implant Elastomer (VIE) tags, and alphanumeric code, which is read by a scanner. Inser- more recently Visible Implant Alphanumeric (VIA) tags, tion is done subcutaneously or in a body cavity (Ferner,
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 130 Andrea Caballero-Gini et alii
2010). However, loss of tags and deleterious effects relat- ed to survival have been detected using this technique (Scherer et al., 2005; Guimaraes et al., 2014). VIE tagging is done through the subcutaneous injection of an elas- tomer mixed with a curing agent. This technique allows the individualization of many animals by creating differ- ent fluorescent color codes as well as placing the marks in various body locations (Moosman and Moosman, 2006). When comparing these three techniques, Branelly et al. (2014) mention that the least efficient is the VIE tagging, due to the migration, darkening and in some cases expul- sion of the tag. VIA tags are compacted elastomers in rectangular shape with an alphanumeric code in fluorescent colors Fig. 1. Location of the VIA tag in the inner thigh of Scinax nasicus. visible under UV light (Heard et al., 2008). The insertion VIA tag was injected just below the skin. is done with an injector provided by the manufacturer, but in some cases the results are better making a previous incision (Buchan et al., 2005; Gower et al., 2006; Heard et toyo Absolute AOS Digimatic. The VIA tags used were al., 2008; Clemas et al., 2009; Kaiser et al., 2009). Despite of standard size (1.2 × 2.7 mm) in fluorescent green with being considered a reliable technique due to the capacity black letters. The insertion site was the ventral region of of individual identification, time of marking, handling of the right thigh following Buchan et al. (2005) (Fig. 1). the individuals, durability of the marks and relatively low The implantation site was sterilized with ethanol 90%, cost (Haw et al., 1990; Buchan et al., 2005; Gower et al., then the tag was inserted under the skin using the injec- 2006), this technique also has disadvantages when used tor provided by NMT following the manufacturer proto- in amphibians: difficulty of tag insertion in species with col and no veterinary glue was used. All marking equip- loose skin, variation on tag retention among species and ment was sterilized between frogs by immersion in 90% darkening of the tag when marking heavily pigmented ethanol for several minutes. Toe-clipping was carried species (Kaiser et al., 2009). Additionally, tag reten- out as control following the method of Martof (1953). tion was low in studies of mark-recapture of tadpoles As suggested by Kinkead et al. (2006), no anesthesia was (Courtois et al., 2013). used in either both procedures. Frogs showed no signs We tested the VIA tags in the lesser snouted treefrog of discomfort, e.g., emitting distress calls or abnormal Scinax nasicus (Cope, 1862) in order to 1) evaluate the movements of the affected limb and/or foot. Frogs were effectiveness of this technique in the identification of put under observation for 24 hours in plastic bags filled individuals, 2) determine the prevalence of tags in mark- with air to observe presence of redness, edema or bleed- recapture studies, and 3) provide some recommendations ing on the treated foot or at the injection site. Subse- for further studies upon this species and others with sim- quently they were released on the same PVC pipe where ilar morphology. they were captured. We marked 196 individuals of S. S. nasicus is a small sized hylid frog distributed in nasicus (55 females, 46 males, and 95 juveniles). Frogs northern and central Argentina, Paraguay, Uruguay, averaged 25.9 mm (SD = 5.34 mm) of SVL. We obtained eastern Bolivia, and central and southern Brazil (Frost, a total of 57 (29%) recaptures, of which 2 individuals 2019). The species is commonly found in open areas of were recaptured three times, 14 individuals twice, and the Atlantic Forest, Cerrado, Chaco, Pampa, and Pantanal 41 individuals only once. The recapture rates in juve- domains (Dalmolin et al., 2017). We captured individuals niles was 47%, males 16% and females 37% (Fig. 2). We of S. nasicus using 173 PVC pipes as a refuge in an area did not find significant differences in the recapture rates of approximately 3 km2 of wetland and associated ripar- of juveniles and adults (t = 0.3487, d.f. = 4, P = 0.7449); ian forest in Benjamín Aceval, Presidente Hayes depart- therefore, it can be suggested that the VIA implants did ment in Paraguay (-24.960522S, -57.359425W). Field not have negative effects on the survival of juveniles. work was done during the months of November 2017, Every time the label was observed, the code could January, March and May 2018. be identified without ambiguity with the help of the UV The individuals were tagged both with VIA tags and light lantern provided by the kit, in 98% of the times. In toe-clipping. Additionally, measurements of the snout- one case, the label was located too deep under the skin vent length (SVL) were taken with a digital caliper Mitu- and could not be read properly. At the same time, by Jan- uary 2018, we observed tissue regeneration in recaptured Evaluating VIA tags in Scinax nasicus 131
Fig. 2. Number of capture and recapture males, females and juve- Fig. 3. Toe pad regeneration in Scinax nasicus. A: after a week of niles of Scinax nasicus. clipping, B: after two months of clipping. frogs where the beginning of the growth of the adhesive movement of tags in dorsal regions of the body is less discs was noticeable, and by March 2018 the discs had common, this area tends to be more pigmented, making similar diameters to those that were not clipped (Fig. reading difficult (Moosman and Moosman, 2006). We 3). In general, discs presented similar shape and colora- opted for the ventral thigh location because this area is tion to those not clipped and no cases of aberrant growth not heavily pigmented in S. nasicus, being almost trans- were observed, such as those described by Hoffman et al. lucid, thus we did not observe any noticeable migra- (2008). tion, which may arise when the target species is large Kaiser et al. (2009) studied the possibility of using and allows movements of the tag in the interfemoral sac VIA tags as a marking method that does not require the (Clemas et al., 2009). We also consider that the use of recapture of animals (by placing the tags on the back of veterinary glue is not necessary since only in one case we a small frog species). They concluded that it is not pos- observed the expulsion of the tag. When we purchased sible to read the codes on the labels without having the the product, manufacturers mentioned that the injection animal in hand. When comparing the handling times needle was re-designed and that it should not be neces- between VIA tags and toe-clipping, Clemas et al. (2009) sary to make an initial incision. However, for marking found that toe-clipping took slightly less amount of initial large or thick-skinned species as well as when marking time of handling and marking than time with the VIA a large number of individuals this may be necessary, in tags, but the last one took less time during identification. order to maintain the instrument sharp and in shape to Despite results showed by Clemas et al. (2009), by using not cause discomfort to the animals. VIA tags in adults and juveniles, we had success identify- The estimation of population and demographic ing individuals of S. nasicus through this methodology. parameters in ecological studies are based on assump- Regarding the second objective of this work, we tions that depend on the marking technique; these are: observed the retention of tags on 97% of the cases. Only (1) no loss of marks; (2) no misidentification of marks; on one occasion the label was expelled and, in another and (3) marking procedures do not alter survival or cap- opportunity, as a result of an inflammation caused in ture probabilities (Seber, 1986; Pollock et al., 1990). This a frog’s leg, the tag had to be removed when the animal last assumption is the most controversial because in most was recaptured. Also, in two cases the tags were invert- techniques’ negative effects on both survival and recap- ed but could be easily rotated by prodding with a finger, ture probability have been observed, being toe-clipping and on one occasion a small piece of the tag broke when the most deleterious technique and VIE and VIA tagging removed from the tag block but was not discarded. the least, although this last one has been scarcely stud- When marking individuals at the caudal end of the ied (Heard et al., 2008; Schmidt and Schwarzkopf, 2010; dorsum, Kaiser et al. (2009) noted that very often the Sapsford et al., 2014, 2015). labels were turned over and migrated ventrally, but they VIA tags are an interesting method to test in do not mention if this affects frog’s survivor or movement amphibians due to its relatively low cost, lower invasive- capacity. They also mention that when tags were not eas- ness when compared to other techniques such as toe- ily detected they remarked animals, and this constituted clipping, and straightforward code interpretation. In our an extra effort and a waste of resources. Contrarily, Heart study, the rate of success of the VIA tag method sug- et al. (2008) compared tag migration placing them in gests that it is suitable for S. nasicus, as it was easy, safe, the thigh and dorsolateral region of the thorax, finding rapid and effective to carry out, as well as easy to detect. that tag retention was higher in the latter site. Although, Moreover, the method is well advised for the species 132 Andrea Caballero-Gini et alii if long term studies are made, due to the pad regenera- (2017): First record of the Lesser Snouted Treefrog tion observed in clipped toes. Still, we suggest keeping a Scinax nasicus (Cope, 1862) in Brazilian coast and hybrid marking method (i.e., VIA tags + toe-clipping) new species records for the state of Rio Grande do when working with other species than S. nasicus, since Sul. Braz. J. Biol. 77: 659-661. VIA tags can have a distinct rate of success depending on Donnelly, M.A., Guyer, C., Juterbock, J.E., Alford, R.A. several factors, such as the degree of stretch in the skin, (1994): Techniques for marking amphibians. In: size of the leg or other body part, transparency. We also Measuring and Monitoring Biological Diversity: advise to take time to inspect carefully the animals to Standard Methods for Amphibians, pp. 277-284. detect VIA tag migration or loss, so there is no resource Heyer, W.R., Donnelly, M.A., McDiarmid, R., Hayek, waste and unnecessary animal stress. L.A.C., Foster M.S., Eds, Smithsonian Institution Press, Washington, DC, USA. Ferner, J.W. (2007): A review of marking and individual ACKNOWLEDGEMENTS recognition techniques for amphibian and reptiles. Herpetological Circular 35, Society for the Study of The permits to perform this study were obtained Amphibians and Reptiles, Atlanta. from the Ministerio del Ambiente y Desarrollo Sosteni- Ferner, J.W. (2010): Measuring and marking post-met- ble (Scientific collection permit N° 173-2017). This work amorphic amphibians. In: Amphibian Ecology and is part of the project PINV15-143 financed by the Con- Conservation: A Handbook of Techniques, pp. 123- sejo Nacional de Ciencia y Tecnología (CONACYT). We 141. Dodd, C.K. Jr., Ed, Oxford University Press, USA. thank for the help received by the staff of Estancia Play- Frost, D.R. (2019): Amphibian Species of the World: An ada and the American Chaco Research Center, also to Online Reference. Version 6.0 (Date of access). Elec- Paloma Moreno and Humberto Sánchez for their help in tronic Database accessible at http://research.amnh. fieldwork. We thank the Programa Nacional de Incentivo org/herpetology/amphibia/index.html. American a los Investigadores (PRONII) from CONACYT. Museum of Natural History, New York, USA. Gower, D.J., Oommen, O.V., Wilkinson, M. (2006): Mark- ing amphibians with alpha fluorescent tags: Caecilians REFERENCES lead the way. Herpetol. Rev. 37: 302. Guimaraes, M., Correa, D.T., Filho, S.S., Oliveira, T.A.L., Brannelly, L.A., Berger, L., Skerratt, L.F. (2014): Compari- Doherty, P.F.J. Sawaya, R.J. (2014): One step forward: son of three widely used marking techniques for adult contrasting the effects of Toe clipping and PIT tagging anuran species Litoria verreauxii alpine. Herpetol. on frog survival and recapture probability. Ecol. Evol. Conserv. Biol. 9: 428-435. 4: 1480-1490. Buchan, A., Sun, L., Wagner, R.S. (2005): Using alpha- Haw, F., Bergman, P.K., Fralick, R.D., Buckley, R.M., numeric fluorescent tags for individual identification Blankenship, H.L. (1990): Visible implanted fish tag. of amphibians. Herpetol. Rev. 36: 43-44. Am. Fish. Soc. Symp. 7: 311-315. Campbell, T.S., Irvin, P., Campbell, K.R., Hoffmann, K., Heard, G.W., Scroggie, M.P. Malone, B. (2008): Visible Dykes, M.E., Harding, A.J., Johnson, S.A. (2009): implant alphanumeric tags as an alternative to toe- Evaluation of a new technique for marking anurans. clipping for marking amphibians: a case study. Wild. Appl. Herpetol. 6: 247-256. Res. 35: 747-759. Clarke, R.D. (1972): The effect of toe clipping on survival Hoffmann, K.E., McGarrity, M.E., Johnson, S.A. (2008): in Fowler’s toad (Bufo woodhousei fowleri). Copeia: Technology meets tradition: A combined VIE-C tech- 182-185. nique for individually marking anurans. Appl. Herpe- Clemas, R.J., Germano, J.M., Speare, R., Bishop, P.J. tol. 5: 265-280. (2009): Use of three individual marking methods in Kaiser, K., Alloush, M., Jones, R.M., Marczak, S., Mar- Australian frogs (Genus: Litoria) with notes on place- tineau, K., Oliva, M. (2009): Use of visual implant ment of Visible Implant Alphanumeric tags. New Alpha (VIAlpha) fluorescent Tags in a small hylid frog Zeal. J. Sci. 34: 1-7. with a new technique for application. Herpetol. Rev. Courtois, E.A., Lelong, C., Calvez, O., Loyau, A., Schmel- 40: 421-422. ler, D.S. (2013): The use of visible implant alpha tags Kinkead, K.E., Lanham, J.D., Montanucci, R.R. (2006): for anuran tadpoles. Herpetol. Rev. 44: 230-233. Comparison of anesthesia and marking techniques on Dalmolin, D.A., Rosa, F.O., Freire, M.D., Fonte, L.F.M., stress and behavioral responses in two Desmognathus Machado, I.F., Paula, C.N., Loebmann, D., Périco, E. Salamanders. J. Herpetol. 40: 323-328. Evaluating VIA tags in Scinax nasicus 133
Martof, B.S. (1953): Territoriality in the green frog, Rana clamitans. Ecology 34: 165-174. McCarthy, M.A., Parris, K.M. (2004): Clarifying the effect of toe clipping on frogs with Bayesian statistics. J. Appl. Ecol. 41: 780-786. Moosman, D.L., Moosman, P.R.J. (2006): Subcutaneous movements of Visible Implant Elastomers in wood frogs (Rana sylvatica). Herpetol. Rev. 37: 300-301. Osbourn, M.S., Hocking, D.J., Conner, C.A., Peterman, W.E., Semlitsch, R.D. (2011): Use of fluorescent visible implant alphanumeric tags to individually mark juve- nile Ambystomatid Salamanders. Herpetol. Rev. 42: 43-47. Pollock, K.H., Nichols, J.D., Brownie, C., Hines, J.E. (1990). Statistical inference for capture recapture experiments. Wild. Monogr. 107: 1-97. Scherer, R.D., Muths, E., Noon, B.R., Corn, P.S. (2005): An evaluation of weather and disease as causes of decline in two populations of boreal toads. Ecol. Appl. 15: 2150-2160. Seber, G.A.F. (1982): The estimation of animal abundance and related parameters. Macmillan, New York. Seber, G.A.F. (1986): A review of estimating animal abun- dance. Biometrics 42: 267-292. Ursprung, E., Ringler, M., Jehle, R., Hödl, W. (2011): Toe regeneration in the neotropical frog Allobates femora- lis. Herpetol. J. 21: 83-86. Waddle, J.H., Rice, K.G., Mazzotti, F.J., Percival, H.F. (2008): Modeling the effect of toe clipping on treefrog survival: beyond the return rate. J. Herpetol. 42: 467- 473.
Acta Herpetologica 14(2): 135-139, 2019 DOI: 10.13128/a_h-7752
Morphological variation of the newly confirmed population of the Javelin sand boa, Eryx jaculus (Linnaeus, 1758) (Serpentes, Erycidae) in Sicily, Italy
Francesco P. Faraone1,*, Salvatore Russotto2, Salvatore A. Barra3, Roberto Chiara3, Gabriele Giacalone4, Mario Lo Valvo3 1 Viale Regione Siciliana S.E., 532, 90129 Palermo, Italy 2 Contrada Grassura Mollaka Faia, s.n., 92027 Licata (AG), Italy 3 Dipartimento Scienze e Tecnologie Biologiche, Chimiche e Farmaceutiche, University of Palermo, Via Archirafi, 18, 90123 Palermo, Italy 4 Cooperativa Silene, Via D’Ondes Reggio, 8/a, 90127 Palermo, Italy *Corresponding author. Email: [email protected]
Submitted on: 2019, 13th May; revised on: 2019, 20th June; accepted on: 2019, 29th June Editor: Dario Ottonello
Abstract. The presence of the Javelin sand boa in Sicily has recently been confirmed. Here the morphological char- acters and sexual dimorphism of the Sicilian population of Eryx jaculus are presented. Seven meristic and six metric characters in 96 specimens from Sicily were examined. The results show that tail length, snout-vent length, the dis- tance between nostrils and the number of ventral and subcaudal scales are different between sexes. The characters found in the Sicilian population of the Javelin sand boa resemble those of the African population (ssp. jaculus) rather than the Eurasian population (ssp. turcicus), but biomolecular studies are necessary to understand its taxonomic iden- tity.
Keywords. Eryx jaculus, Serpentes, morphological variation, folidosis.
The Javelin sand boa Eryx jaculus (Linnaeus, 1758), Recently, the presence of E. jaculus has been con- a member of the family Erycidae, is found in the south- firmed in a small area of southern Sicily, in Italy (Insacco ern Balkans, Middle East and North Africa (Sindaco et et al., 2015). Knowledge of the Javelin sand boa in Sicily al., 2013). Two morphologically defined subspecies are is currently limited to scant preliminary information on usually recognized for this species. The nominate sub- the population’s distribution, morphology, habitat and species is present in North Africa and the ssp. turcicus diet (Insacco et al., 2015; Faraone et al., 2017a, b). (Olivier, 1801) is found in the Balkans and the Mid- In this paper the first comprehensive report of the dle East (Tokar and Obst, 1993; Sindaco et al., 2013; morphological characters and sexual dimorphism in the Geniez, 2015). According to Tokar and Obst (1993), Sicilian population of E. jaculus is presented. the Transcaucasian ssp. familiaris Eichwald, 1831 Sampling was carried out within the currently known could be synonymous with E. j. turcicus. Morphologi- geographic range of the Javelin sand boa in South Central cal variations of E. jaculus have been reported both at Sicily, within an area between Palma di Montechiaro (prov- a global (Tokar, 1991) and a local scale (Cattaneo, 1984, ince of Agrigento) and Butera (province of Caltanissetta) 2005, 2010; Rhadi et al., 2015; Zarrintab et al., 2017; (see Insacco et al., 2015; Faraone et al., 2017b). A total of Eskandarzadeh et al., 2018). 55 adult (30 males; 25 females) and 41 juvenile (21 males;
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 136 Francesco P. Faraone et alii
b 17 females; 3 undetermined) specimens were examined Z =Yi (SVL/ SVLi ) between August 2014 and September 2018. Among them, 47 specimens were collected during active searches at night, following Lleonart et al. (2000), where Z is the trans- 12 specimens were rescued from abandoned cisterns and formed value of the variable Y, which is the variable the tunnel greenhouse in which they were trapped, four affected by size, represented as the snout-vent length were found dead inside cisterns, one had been killed by (SVL), and b is the slope of the linear regression between domestic cats and 32 specimens were roadkilled. logY and logSVL. The data were checked for parametric Sex was determined by examining external sexual assumptions using Levene’s test and the Shapiro-Wilk features (spurs, tail shape) and, mainly in young snakes, test. For each variable, parametric (independent t-test) also by cloacal popping. The reliability of external fea- or non-parametric (Mann-Whitney U test) analysis was tures for sexual determination was also confirmed by conducted, with a level of significance at 0.05. Differ- cloacal probing in dead specimens. In absence of detailed ences between the sexes were checked by using the Prin- references, E. jaculus with a snout-vent length < 270 mm cipal Component Analysis (PCA), applied to the cor- (see Itescu et al., 2018; Eskandarzadeh et al., 2018) were relation matrix. Body weight (W) was not included in considered as juveniles. PCA, because its value can vary greatly depending on the Seven meristic characters were considered: the num- condition of each specimen (pregnancy, decay, freshly ber of scales between the eyes (BE), around the eye (RE), ingested prey, etc.). SVL and the characters with a P value posterior to the internasal (PIN), between the eye and the greater than 0.05 were also excluded. nasal (BEN), the rows of scales between the eye and the Color morph was classified according to the three supralabial area (BES), the ventrals (VS) and the subcaudals types reported by Tokar (1991) and Tokar and Obst (ScdS). Six metric characters were also recorded: snout-vent (1993). The ‘discrete’ morph is characterized by a reduced length (SVL), tail length (TL), the distance between the dorsal pattern, with narrow transverse dorsal bars and eyes (DBE), the distance between the posterior edge of the small scattered spots on the sides. In the ‘standard’ eye and the tip of the snout (DES), the distance between morph, the dorsal pattern is more extensive than in the the nostrils (DBN) as well as body weight (W). For bilat- ‘discrete’ morph and the dorsal blotches are spaced by eral characters, only the right side was recorded. Biometric light areas of similar thickness and are not in contact characters were measured using a digital caliper accurate to with each other. The ‘negative’ morph is characterized by 0.01 mm, except for SVL, which was measured with a mm a greater development of the dark pattern, while the light ruler, and body weight (W), which was measured only on parts are reduced to narrow suboval blotches on the back. live individuals using a digital scale with 1 g precision. The descriptive statistic and a comparison between In order to remove the information related to size, sexes were made for adults (Table 1). Significant inter- each metric variable was transformed using the formula: sexual differences in five characters were found: the num-
Table 1. Descriptive statistics and univariate comparisons between sexes of the Sicilian population of Eryx jaculus. The first five characters are in millimetres, the sixth in grams and the others are meristic. Significant differences (P<0.05) are shown in bold. *= transformed values.
Males Females T U P Mean ± SD Range n Mean ± SD Range n SVL 342.1 ± 42.4 270.0 - 420.0 26 400.5 ± 86.1 270.0 - 580.0 23 3.2 - 0.004 *TL 40.7 ± 6.0 26.9 - 53.9 26 31.4 ± 6.2 20.7 - 45.9 23 7.9 - 0.000 *DES 7.1 ± 0.7 5.8 - 8.7 26 7.6 ± 1.2 5.9 - 10.5 21 1.8 - 0.086 *DBN 4.0 ± 0.4 3.3 - 4.7 26 4.2 ± 0.6 3.3 - 5.6 21 2.4 - 0.019 *DBE 7.0 ± 0.7 6.1 - 8.3 26 7.5 ± 1.1 5.5 - 10.0 21 - 165.0 0.194 W 36.9 ± 19.5 18 - 95 14 64.2 ± 49.6 15 – 204 16 - 76.0 0.135 VS 175.9 ± 2.8 170 - 182 26 181.9 ± 3.9 176 - 190 21 6.1 - 0.000 ScdS 25.2 ± 1.9 22 - 29 29 19.7 ± 1.4 17 - 22 23 11.8 - 0.000 PIN 2.6 ± 0.5 2 - 3 30 2.6 ± 0.5 2 - 3 24 - 357.0 0.958 BE 6.8 ± 0.6 6 - 8 30 6.8 ± 0.4 6 - 7 24 - 358.0 0.972 RE 9.4 ± 0.8 8 - 11 29 9.0 ± 0.7 8 - 10 22 - 238.5 0.126 BEN 3.0 ± 0 3 - 3 30 3.0 ± 0 3 - 3 24 - - - BES 1.0 ± 0.2 1 - 2 30 1.0 ± 0 1 - 1 24 - 348.0 0.835 Morphological variation of Eryx jaculus in Sicily 137
Fig. 1. Scatterplot of scores with the equiprobability ellipses at 95% projected on the first two principal components between sexes of Fig. 2. Examples of color and pattern variation in the Sicilian popu- the Sicilian population of Eryx jaculus. lation of Eryx jaculus. (A) Adult, male (B) Adult, female (C) Juve- nile, male (D) Ventral variation in three adult females. ber of ventral scales and snout-vent length have higher the ‘negative’ phenotype, but some small parts of their values in females and the number of subcaudal scales, back looked like the ‘standard’ morph, similar to what the distance between the nostrils and the tail length have was observed in some E. jaculus by Tokar and Obst higher values in males. The result of the Principal Com- (1993) and Werner (2016). Young individuals were gener- ponents Analysis (PCA) shows that 85.1% of the vari- ally lighter than adults, with a lower contrast in the dark ance is explained by the first two components (PC1 = pattern. No difference in dorsal color pattern was noticed 58.1%, PC2 = 27.0%). The scatterplot of scores with the between sexes. The dorsal base color is orange, sand or equiprobability ellipses at 95% (Lagonegro and Feoli, yellowish. The ventral, paraventral and some adjacent 1985) projected on the first two principal components scales were usually pale yellowish orange (in only one (Fig. 1) shows a clear separation between sexes on PC1, large female it was a whitish-cream color). Ventral parts with males distributed within the negative values of the were adorned with small blackish spots. The orange color axis and the females within the positive values. of the belly in a few cases extended to the subcaudal The correlation coefficients of the variables (Table 2) scales, which usually appeared whitish with a few scat- show that PC1 has a strong positive correlation with ven- tered orange scales. tral scales (VS) and a negative correlation with subcaudal The results show clear intersexual differences in the scales (ScdS) and tail length (TL). Sicilian population of E. jaculus. The females usually The examined specimens showed different color pat- reach a greater size in terms of snout-vent length and terns (Fig. 2). In most of the snakes, the ‘standard’ morph have a greater number of ventral scales than males, and was identified, however in some individuals the dorsal males have a greater number of subcaudal scales, a longer dark blotches were so extended that they appeared simi- tail and a greater distance between the nostrils. Signifi- lar to the ‘negative’ morph, yet these spots were not con- cant intersexual differences in ventral scales have already nected on the sides, so they were assigned to the ‘stand- been observed in this species (Tokar, 1991) but are not ard’ type. Seven snakes could generally be attributed to found in some Asian populations (Rhadi et al., 2015; Eskandarzadeh et al., 2018) nor in some other Eryx spe- cies (Al-Sadoon and Al-Otaibi, 2014; Eskandarzadeh et Table 2. Correlation coefficients of the variables for the first two al., 2013). factors used in the Principal Component Analysis (PCA). The pattern of sexual dimorphism observed in the Variables PC 1 PC 2 Sicilian population is also reported in other species of snakes (Marques et al., 2006, Pizzatto et al., 2007, Zanella TL -0,81 0,52 and Cechin, 2010, Siqueira et al., 2013, Manjarrez et al., DBN -0,40 -0,88 2014). Larger females can provide more space for incu- VS 0,81 0,16 bating eggs and embryos and more body reserves to ScdS -0,92 0,07 138 Francesco P. Faraone et alii invest in reproduction (Bonnett et al., 1998) and gener- tively (see also Tokar, 1991). The morphological affinities ally, have a larger clutch size (Shine, 1993) and lower between the Sicilian and African populations reported mortality in their offspring (Kissner and Weatherhead, here has only a preliminary value and, obviously, a bio- 2005). Moreover, in various snake species the number of molecular approach is necessary in order to properly ventral scales corresponds to the number of dorsal ver- assess the phylogeographic and taxonomic structure of tebrae, which are generally more numerous in the larger the Javelin sand boa. sex (Shine, 2000). On the other hand, smaller males have greater agility during the search for females and a lower metabolic cost (Rivas and Burghardt, 2001). A larger tail ACKNOWLEDGEMENTS and a greater number of subcaudal scales in males are linked to the positioning of the hemipenes within the tail We are grateful to Agostino Cantavenera, Matteo and produce better performance in the courtship phase Di Nicola, Angelica Rallo, Alex Venutelli for their help (King, 1989; Luiselli, 1996; Shine et al., 1999). The males in fieldwork. We also thank Edoardo Razzetti for kindly of Sicilian E. jaculus also have a relatively larger interna- providing us some interesting bibliographic references sal distance (DBN) than the females. Sexual dimorphism and Viviana Tinnirello for language revision. The tem- in head shape is a phenomenon known in various snake porary capture and handling have been carried out with species and is often linked to a different trophic niche the permits established by law (MATT prot. N.2766 / between males and females (Camilleri and Shine, 1990; T-A31, 12/01/2018 and Regione Siciliana Prot. N, 1637, Vincent et al., 2004). This result deserves to be investi- 24/01/2018). gated using appropriate tools (Geometric morphometric analysis) and its relation to the dietary habits of this pop- ulation, which is currently being studied, should be taken REFERENCES into account. The three dorsal patterns proposed by Tokar (1991) Al-Sadoon, M.K., Al-Otaibi, F.S. (2014): Ecology of the should be used with caution. It is sometimes difficult to sand boa, Eryx jayakari in Riyadh Region of Saudi apply these patterns to specimens with mixed phenotypes Arabia. Saudi J. Biol. Sci. 21: 391-393. (see Tokar and Obst, 1993) that may be locally common Bonnet, X., Shine, R., Naulleau, G., Vacher-Vallas, M. (Werner, 2016). (1998): Sexual dimorphism in snakes: Different repro- The subspecies of E. jaculus, proposed by Tokar and ductive roles favour different body plans. Proc. R. Soc. Obst (1993), is currently accepted (Sindaco et al., 2013; Lond. B. Biol. Sci. 265: 179-183. Werner, 2016). The criteria proposed by Tokar and Obst Camilleri, C., Shine, R. (1990): Sexual dimorphism and (1993) are based on an in-depth revision of the mor- dietary divergence: differences in trophic morphology phological variation of E. jaculus within its geographical between male and female snakes. Copeia 3: 649-658. range (Tokar, 1991) and have been confirmed by recent Cattaneo, A. (1984): Podarcis erhardii naxensis ad research on a local scale (Eskandarzadeh et al., 2018; Antiparos (Cicladi centrali) e note di campagna Rhadi et al., 2015; Zarrintab et al., 2017). sull’erpetocenosi dell’isola (Reptilia). Atti Soc. ital. Sci. Comparing the characters reported by Tokar and nat. Mus. civ. Stor. nat. Milano 125: 245-254. Obst (1993) with those of the examined specimens in Cattaneo, A. (2005): Osservazioni sull’erpetofauna this study, a higher resemblance between the Sicilian dell’isola greca di Kos (Sporadi meridionali) con un population and the North African nominal subspecies, inedito caso di simpatria microinsulare fra due specie rather than with the Eurasian ssp. turcicus, was detect- affini di Colubridi: Hierophis caspius (Gmelin) e Hier- ed. In the Sicilian population, as well as in E. j. jaculus, ophis jugularis (L.). Atti Mus. Stor. nat. Maremma, there are usually seven scales between the eyes (74.7%; Grosseto 21: 79-91. n = 95) (usually six in E. j. turcicus) and three postint- Cattaneo, A. (2010): Note eco-morfologiche su alcune ernasal scales (64.6%; n = 96) (usually two in E. j. turci- specie ofidiche egee, con particolare riferimento alle cus). Moreover, the Sicilian ventral scales range (170-190) popolazioni delle Cicladi centro-orientali (Reptilia). overlaps with both the nominal subspecies (170-198) and Naturalista sicil., 34: 319-350. ssp. turcicus (161-200). Eskandarzadeh, N., Darvish, J., Rastegar-Pouyani, E., Furthermore, a more frequent presence of the ‘stand- Ghassemzadeh, F. (2013): Reevaluation of the taxo- ard’ phenotype (92.7%; n = 96) is observed with respect nomic status of sand boas of the genus Eryx (Daudin, to the ‘negative’ and ‘discrete’ dorsal patterns, which are 1803) (Serpentes: Boidae) in northeastern Iran. Turk. very common in European and Asian populations respec- J. Zool., 37: 1-9. Morphological variation of Eryx jaculus in Sicily 139
Eskandarzadeh, N., Rastegar-Pouyani, N., Rastegar-Pouy- In: Herpetologia no Brasil II, p. 201-221. Nascimento, ani, E., Todehdehghan, F., Rajabizadeh, M. (2018): L.B., Oliveira, M.E., Eds, Sociedade Brasileira de Her- Sexual Dimorphism in the Javelin Sand Boa, Eryx petologia, Belo Horizonte. jaculus (Linnaeus, 1758) (Serpentes: Erycidae), from Rhadi, F.A., Rastegar-Pouyani, N., Karamiani, R., Western Iran. Curr. Herpetol. 37: 88-92. Mohammed, R.G. (2015): Taxonomic status of sand Faraone, F.P., Barra, S.A., Giacalone, G., Chiara, R., Rus- boas of the genus Eryx (Daudin, 1803) (Serpentes: sotto, S., Lo Valvo, M. (2017a): First observations of Boidae) in Bahr Al-Najaf Depression, Al-Najaf Prov- oophagy in a wild population of the sand boa (Eryx ince, Iraq. Iran. J. Anim. Biosyst. 11: 149-156. jaculus). Herpetol. Bull. 142: 48-49. Rivas, J.A., Boughardt, G.M. (2001): Understanding sex- Faraone, F.P., Chiara, R., Barra, S.A., Giacalone, G., Lo ual size dimorphism in snakes: wearing the snake’s Valvo, M. (2017b): Nuovi dati sulla presenza di Eryx shoes. Anim. Behav. 62: F1-F6. jaculus (Linnaeus, 1758) in Sicilia. In: Atti XI Con- Shine, R., (1993): Sexual Dimorphism in Snakes. In: gresso Nazionale della Societas Herpetologica Italica, Snakes: Ecology & Behavior, pp. 49-86. Seigel, R.A., Trento 2016, pp. 75-79. Menegon, M., Rodriguez-Pri- Collins J.T., Eds, McGrill-Hill, New York. eto, A., Deflorian, M.C., Eds, Ianieri Edizioni, Pescara. Shine, R. (2000): Vertebral numbers in male and female Geniez, P. (2015): Serpents d’Europe, d’Afrique du snakes: the roles of natural, sexual and fecundity Nord et du Moyen-Orient. Guide Delachaux. Paris, selection. J. Evolution. Biol. 13:455-465. Delachaux et Niestlé S.A. Shine, R.G., Olsson, M.M., Moore, I.T., LeMaster, M.P., Insacco, G., Spadola, F., Russotto, S., Scaravelli, D. (2015): Mason, R.T. (1999): Why do male snakes have long- Eryx jaculus (Linnaeus, 1758): A new species for the er tails than females? Proc. R. Soc. Lond. B. Biol. Sci. Italian herpetofauna (Squamata: Erycidae). Acta Her- 266: 2147-2151. petol. 10: 149-153. Sindaco, R., Venchi, A., Grieco, C. (2013): The Reptiles Itescu, Y., Schwarz, R., Donihue, C.M., Slavenko, A., of the Western Palearctic. 2. Annotated checklist and Roussos, S.A., Sagonas, K., Valakos E.D., Foufopou- distributional atlas of the snakes of Europe, North los J., Pafilis P., Meiri, S. (2018): Inconsistent patterns Africa. Middle East and Central Asia, with an update of body size evolution in co‐occurring island reptiles. to the, Vol 1, Edizioni Belvedere, Latina. Glob. Ecol. Biogeogr. 27: 538-550. Siqueira, D.M., Nascimento, L.P., Montingelli, G.G., San- King, R.B. (1989): Sexual dimorphism in snake tail tos-Costa, M.C. (2013): Geographical variation in length: sexual selection, natural selection, or morpho- the reproduction and sexual dimorphism of the Bod- logical constraint? Biol. J. Linn. Soc. 38: 133-154. daert’s tropical racer, Mastigodryas boddaerti (Ser- Kissner, K.J., Weatherhead, P.J. (2005): Phenotypic effects pentes: Colubridae). Zoologia 30: 475-481. on survival of neonatal Northern Watersnakes Nero- Tokar, A.A. (1991): A revision of the subspecies struc- dia sipedon. J. Anim. Ecol. 74: 259-265. ture of Javelin sand boa, Eryx jaculus (Linnaeus, 1758) Lagonegro, M., Feoli, E. (1985): The use of ellipses of (Reptilia, Boidae). Herpetological Researches 1: 18-41. equal concentration to analyse ordination vegetation Tokar, A., Obst, F.J. (1993): Eryx jaculus - Westliche patterns. Studia Geobot. 5: 143-165. Sandboa. In: Handbuch der Reptilien und Amphibi- Lleonart, J., Salat, J., Torres, G.J. (2000): Removing allo- en Europas, Band 3/I., Schlangen (Serpentes) I, pp. metric effects of body size in morphological analysis. 35-54. Böhme,W., Ed., Aula-Verlag, Wiesbaden. J. Theoret. Biol. 205: 85-93. Vincent, S.E., Herrel A., Irschick, D.J. (2004): Sexual dimor- Luiselli, L. (1996): Individual success in mating balls of phism in head shape and diet in the cottonmouth snake the grass snake, Natrix natrix: size is important. J. (Agkistrodon piscivorus). J. Zool., Lond. 264: 53-59. Zool. 239: 731-740. Werner, Y.L. (2016): Reptile life in the land of Israel. Edi- Manjarrez, J., Contreras-Garduño J., Janczur M.K. (2014): tion Chimaira, Frankfurt am Main. Sexual size dimorphism, diet, and reproduction in the Zanella, N., Cechin, S.Z. (2010): Reproductive biology of Mexican garter snake, Thamnophis eques. Herp. Cons. Echinanthera cyanopleura (Serpentes: Dipsadidae) in Biol. 9(1): 163-169. southern Brazil. Zoologia 27: 30-34. Marques, O.A.V., Sawaya, R.J., Stender-Oliveira, F., Zarrintab, M., Milto, K.D., Eskandarzadeh, N., Zangie, Franca, F.G.R. (2006): Ecology of the colubrid snake B., Jahane, M., Gholi Kamif, H., Rastegar-Pouyani, N., Pseudablabes agassizii in southeastern south America. Rastegar-Pouyani, E., Rajabizadeh, M. (2017): Taxon- Herpetol. J. 16: 37-45. omy and distribution of sand boas of the genus Eryx Pizzato, L.P., Almeida-Santos, S.M., Marques, O.A.V. Daudin, 1803 (Serpentes: Erycidae) in Iran. Zool. (2007): Biologia reprodutiva das serpentes brasileiras. Middle East 63: 117-129.
Acta Herpetologica 14(2): 141-145, 2019 DOI: 10.13128/a_h-7753
Variability in the dorsal pattern of the Sardinian grass snake (Natrix natrix cetti) with notes on its ecology
Enrico Lunghi1,2,3,4,*, Simone Giachello5, Manuela Mulargia6, Pier Paolo Dore7, Roberto Cogoni8, Claudia Corti2 1 Institute of Zoology, Chinese Academy of Sciences, Beichen West Road 1, 100101, Beijing, China. *Corresponding author. Email: enri- [email protected] 2 Museo di Storia Naturale dell’Università di Firenze, Sede “La Specola”, Via Romana 17, 50125 Firenze, Italia 3 Natural Oasis, Via di Galceti 141, 59100 Prato, Italia 4 Universität Trier Fachbereich VI Raum-und Umweltwissenschaften Biogeographie, Universitätsring 15, 54286 Trier, Germany 5 Dipartimento di Scienze e politiche ambientali, Università degli Studi di Milano, via Celoria 2, 20133 Milano, Italia 6 CEAS Santa Lucia Siniscola. Via Isalle 4, 08029 Siniscola, Italia 7 Gruppo Speleo Ambientale Sassari, Via Pigliaru, 3, 07100 Sassari, Italia 8 Unione Speleologica Cagliaritana, via Scarlatti, 11, 09045 Quartu Sant’Elena, Italia
Submitted on: 2019, 4th July; revised on: 2019, 2nd September; accepted on: 2019, 17th September Editor: Marco Mangiacotti
Abstract. The Sardinian grass snake (Natrix natrix cetti) is a Critically Endangered snake endemic to Sardinia (Italy), for which information is still scarce. In the present work, we report information obtained from 36 observations of N. n. cetti performed in different areas of the Island. Three different colorations were mainly observed and darker snakes were in general males and big adults; the only juvenile found showed a complete different dorsal colouration. Snakes were observed active during day-time and often far from the aquatic habitats.
Keywords. Melanism, abundism, mimicry, predator avoidance, island.
The Sardinian grass snake Natrix( natrix cetti) is a ogy and biology (Lunghi et al., 2016; Lunghi et al., 2018; Critically Endangered snake endemic to Sardinia, for Vanni and Cimmaruta, 2010). It is known that N. n. cetti which we still have very limited information (European can exploit different environments, from dry rocky areas Reptile & Amphibians Specialist Group, 1996; Vanni and to wetlands, rivers and even caves, and that the species is Cimmaruta, 2010). A recent phylogenetic analysis per- relatively widespread on the Island (Capula et al., 1994; formed on the Natrix natrix complex grouped the Sardin- de Pous et al., 2012; Lanza 1986; Mulargia et al., 2018; ian grass snake together with the Corsican subspecies (N. Salvi and Bombi, 2010). However, the presence of the n. corsa) into a distinct genetic group, which was likely Viperine snake (Natrix maura), as potential competitor, promoted by their isolation occurred during glaciations seems to affect its distribution (Stefani 1983; Vanni and (Fritz et al., 2012; Kindler et al., 2013). According to the Cimmaruta 2010). most recent phylogenetic study, the western grass snakes Adults of N. n. cetti differ morphologically from the belong to a new clade called Natrix helvetica (Kindler et continental subspecies because of the lack of the typi- al., 2017); however, in this study no samples from Sardin- cal light “collar”, and the smaller size (Speybroeck et al., ia have been analysed. Its elusiveness could be one of the 2016; Vanni and Cimmaruta, 2010). Stefani (1983) causes of the few data available on its distribution, ecol- reports for N. n. cetti a typical light greyish background
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 142 Enrico Lunghi et alii colour, a characteristic that should be of help in distin- throughout the year, but the most in spring (Jan-Mar = guishing the Sardinian grass snake from the Corsican 9.1%; Apr-Jun = 70.6%; Jul-Sep = 14.4%; Oct-Dec = 5.9%; one, being the latter characterised by a dark greenish total surveys = 153). For each snake we recorded: local- background. Recently, abundistic (i.e., dorsum character- ity, elevation and time of the observation, habitat typology ized by enlarged dark stripes) and melanotic (i.e., near- and dorsal background colour pattern (greyish, greenish- ly completely black coloration) individuals of N. n. cetti brownish, melanotic) (Fig. 1). Captured snakes were were observed (Lunghi et al., 2016), thus increasing the measured (total length) and sexed. Total length was meas- variability of the dorsal colouration known for this sub- ured photographing the snakes on a plasticised millimetre species. Furthermore, as far as we know, no information paper; snakes’ length was extrapolated using the program exists on juveniles. ImageJ. Sex was assessed combining visual inspection of Here we report observations of Natrix natrix cet- the morphology of the snake (proportion of the head and ti gathered during three years fieldwork. All captured tail) and number of sub-caudal scales (Vanni and Cim- snakes were measured, and information on dorsal pattern maruta, 2010). Head pattern was used for individually and habitat were recorded. The observation of one juve- snake recognition (Sacchi et al., 2016; Vaughan, 1999). A nile is also reported. Generalized Linear Model (R software with nlme package; From 2016 to 2018, we conducted herpetological field Pinheiro et al., 2016; R Development Core Team 2018) observations focusing on Eastern and Southern Sardinia was used to assess whether snake coloration is related to (see Table 1; coordinates are not reported for species safe- sex, elevation, time of survey and total length (TL). We guard, see Lunghi et al., 2019). Repeated linear transects, used colouration as dependent variable, while sex, TL, based on VES (Visual Encounter Survey), were carried time of survey and elevation as independent variables; out. The surveys were performed by day (9 a.m. – 5 p.m.) year and transect identity as random factors. To use col-
Fig. 1. The four different dorsal colourations of Natrix natrix cetti reported in this study: A) greyish, B) greenish-brownish, C) black, D) the juvenile. (Photos A and B by M. Di Nicola; C by E. Lunghi; D by S. Giachello). Dorsal pattern in Sardinian grass snake 143
Table 1. Data of the captured individuals of Natrix natrix cetti: sex, total length (TL), dorsal background colour (1 = greyish, 2 = greenish- brownish, 3 = melanotic), elevation (m a.s.l.), time at which snakes were observed (24h), province, year, transect identity and the mountain complex. *individuals captured twice.
Sex TL (cm) Colour Elevation Time Province Year Transect Mountain Female 65.7 2 341 16:15 Carbonia-Iglesias 2017 Barega1 Barega Female 44 1 474 14:32 Cagliari 2017 Cagliari3 Sette Fratelli Female 72.1 2 515 10:30 Cagliari 2016 Cagliari2 Sette Fratelli Female 74 1 534 10:08 Cagliari 2018 Cagliari2 Sette Fratelli Female 53.1 1 535 11:10 Cagliari 2017 Cagliari2 Sette Fratelli Female 62.3 2 539 12:00 Cagliari 2017 Cagliari2 Sette Fratelli Female 61.8 2 541 14:01 Cagliari 2017 Cagliari2 Sette Fratelli Male 54.3 2 633 11:02 Cagliari 2017 Cagliari2 Sette Fratelli Female 59 2 634 10:36 Cagliari 2017 Cagliari2 Sette Fratelli *Female 63 2 638 16:21 Cagliari 2017 Cagliari2 Sette Fratelli Female 57.9 2 670 15:45 Cagliari 2016 Cagliari2 Sette Fratelli Male 64.7 2 689 14:05 Cagliari 2017 Cagliari2 Sette Fratelli Male 60.1 3 706 13:45 Cagliari 2016 Cagliari2 Sette Fratelli Female 58.8 2 730 16:42 Cagliari 2016 Cagliari2 Sette Fratelli *Female 63 2 736 16:40 Cagliari 2016 Cagliari2 Sette Fratelli Female 63 1 792 15:58 Cagliari 2018 Cagliari2 Sette Fratelli Male 46.4 2 830 15:50 Cagliari 2017 Cagliari2 Sette Fratelli Female 59.8 2 863 13:20 Cagliari 2017 Cagliari3 Sette Fratelli Male 55.5 1 1029 14:45 Nuoro 2018 M_albo3 Monte Albo
ouration as dependent variable, we ascribed an ascending coloration was most frequently observed in the longer order to the different colourations, going from the light- snakes (Fig. 2). No significant effect was observed for est (greyish = 1) to the darkest (greenish = 2) (Table 1). other variables (sex, F1,8 = 2.00, P = 0.195; elevation, F1,8 In this analysis we excluded the juvenile, the single mel- = 2.90, P = 0.127; time of survey, F1,8 = 1.69, P = 0.229). anotic individual, and, for the individual captured twice During our study, a female with a wounded eye was (see below) only the first observation. captured twice, the first on May 2016 and the second in In total we performed 153 surveys on 3 different are- April 2017; in both occasions this female was active at as (Barega, no. of transects = 1, total surveys = 5; Mon- about the same time (16:41 and 16:21 respectively). This te Albo, no. of transects = 11, total surveys = 102; Sette individual was recaptured almost 90 m far from the site Fratelli, no. of transects = 2, total surveys = 46) and we of first encounter (difference in altitude 98 m). observed 19 Natrix natrix cetti; seventeen were adults. The Sardinian grass snake is one of the least studied Most of the observations (17) were performed on the Italian species. Elusiveness, rather than its potential rarity, Sette Fratelli Mountain and only an adult female was cap- could be the reasons for the few existing studies on this tured twice. The greenish-brownish (n = 12) coloration species (Vanni and Cimmaruta, 2010); indeed, although resulted to be the most common, followed by the grey- being made of just 19 records, our dataset is one of the ish (n = 5) and the melanotic (n = 1) (Fig. 1A-C). The richest available. The dorsal coloration of Natrix natrix observed juvenile (TL = 22 cm) showed a complete dif- cetti seems to be more variable than previously reported; ferent dorsal colouration: the background colour was indeed, most of our observations highlighted the high black, the stripes white and a white collar was also pre- frequency of the greenish-brownish coloration, a charac- sent (Fig. 1D). Snakes were observed at an elevation teristic that should be typical of the Corsican grass snake between 341 and 1029 m a.s.l. Two individuals were (Stefani, 1983). found in small temporary water bodies, one in a mine, Because of our dataset paucity, we cannot provide while the others 15 in rocky areas sometimes even more any assumptions on the causes supporting the high pat- than 1 km far from the nearest water body. tern differentiation observed in this subspecies (but see Snakes’ coloration significantly correlated only with also Lunghi et al., 2016). Further studies are needed to total length (F1,8 = 8.94, P = 0.017); the darker (greenish) assess whether factors are influencing the observed vari- 144 Enrico Lunghi et alii
Castella, B., Golay, J., Monney, J.-C., Golay, P., Mebert, K., Dubey, S. (2013): Melanism, body condition and elevational distribution in the asp viper. J. Zool. 290: 273-280. Clusella Trullas, S., van Wyk, J.H., Spotila, J.R. (2007): Thermal melanism in ectotherms. J. Therm. Biol. 32: 235-245. de Pous, P., Speybroeck, J., Bogaerts, S., Pasmans, F., Beu- kema, W. (2012): A contribution to the atlas of the terrestrial herpetofauna of Sardinia. Herpetol. Notes 5: 391-405. European Reptile & Amphibians Specialist Group (1996): Natrix natrix ssp. cetti. The IUCN Red List of Threat- ened Species 1996: e.T14364A4436077. Fritz, U., Corti, C., Päckert, M. (2012): Mitochondrial DNA sequences suggest unexpected phylogenetic Fig. 2. Boxplots indicating differences in Natrix natrix cetti body position of Corso-Sardinian grass snakes (Natrix cetti) size (total length) per dorsal coloration (only for greyish and green- and do not support their species status, with notes on ish background colorations). Individuals used here are the same phylogeography and subspecies delineation of grass used in the GLM analysis (see Table 1). Dark bar inside boxes rep- resents the median, boxes represent the range between 1st and 3rd snakes. Org. Divers. Evol. 12: 71-80. quartile, whiskers represent the variability outside the upper/lower Fulgione, D., Lega, C., Trapanese, M., Buglione, M. quartile (box), circles represent outliers. (2015): Genetic factors implied in melanin-based coloration of the Italian wall lizard. J. Zool. 296: 278-295. ability and, if darker coloration may be beneficial for Kindler, C., Böhme, W., Corti, C., Gvoždík, V., Jablonski, snakes’ longevity (Castella et al., 2013; Clusella Trullas et D., Jandzik, D., Metallinou, M., Široký, P., Fritz, U. al., 2007; Fulgione et al., 2015; Stevens et al., 2009; Zuffi (2013): Mitochondrial phylogeography, contact zones 2008). The juvenile highlighted ontogenetic variability in and taxonomy of grass snakes (Natrix natrix, N. meg- dorsal coloration in N. n. cetti. alocephala). Zool. Scr. 42: 458-472. Our study agrees with Stefani (1983) and Lanza Kindler, C., Chèvre, M., Ursenbacher, S., Böhme, W., (1986) considering N. n. cetti not strictly related to water Hille, A., Jablonski, D., Vamberger, M., Fritz, U. bodies. This likely allows the snake to exploit further hab- (2017): Hybridization patterns in two contact zones itats, increasing prey availability and lowering competi- of grass snakes reveal a new Central European snake tion with Natrix maura (Lunghi et al., 2018; Stefani, 1983; species. Sci. Rep. 7: 7378. Vanni and Cimmaruta, 2010). Our study was carrried out Lanza, B. (1986): I Rettili e gli Anfibi. In: L’ambiente nat- in day-time (see also Lanza, 1986) but we cannot exclude urale in Sardegna (Elementi di base per la conoscenza nocturnal activity as reported by Capula et al. (1994). e la gestione del territorio), pp. 289-321. Camarda, I., Falchi, S. Nudda, G., Eds, Carlo Delfino Editore, Sas- sari. ACKNOWLEDGMENTS Lunghi, E., Corti, C., Manenti, R., Ficetola, G.F. (2019): Consider species specialism when publishing datasets. The study was authorised by the Italian Minis- Nat. Ecol. Evol. 3: 319. try of Environment (n. 68754/T-A31 of 28/11/2016) Lunghi, E., Deschandol, F., Cornago, L., Cogoni, R. and by Regione Autonoma della Sardegna (n° 9112 of (2016): Dark coloration in Sardinian grass snakes 04/05/2017). (Natrix natrix cetti). Herpetol. Bull. 137: 28-29. Lunghi, E., Mascia, C., Mulargia, M., Corti, C. (2018): Is the Sardinian grass snake (Natrix natrix cetti) an REFERENCES active hunter in underground environments? Spixiana 41: 160. Capula, M., Rugiero, L., Luiselli, L. (1994): Ecological Mulargia, M., Corti, C., Lunghi, E. (2018): The herpeto- observations on the Sardinian grass snake, Natrix fauna of the Monte Albo, Sardinia (Italy). Russ. J. natrix cetti. Amphibia-Reptilia 15: 221-227. Herpetol. 25: 172-176. Dorsal pattern in Sardinian grass snake 145
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., Team, R.C. (2016): nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1-128. R Development Core Team (2018): R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http:// www.R-project.org/. [accessed 28 September 2018] Sacchi, R., Scali, S., Mangiacotti, M., Sannolo, M., Zuffi, M.A.L. (2016): Digital identification and analysis. In: Reptile Ecology and Conservation, pp. 59-72. Ken- neth Dodd, C.J., Eds, Oxford University Press, Oxford. Salvi, D., Bombi, P. (2010): Reptiles of Sardinia: updating the knowledge on their distribution. Acta Herpetol. 5: 161-177. Speybroeck, J., Beukema, W., Bok, B., Van Der Voort, J., Velikov, I. (2016): Field guide to the Amphibians & Reptiles of Britain and Europe, Bloomsbury, London. Stefani, R. (1983): La Natrice del Cetti. Biogeographia 8: 745-755. Stevens, M., Winney, I.S., Cantor, A., Graham, J. (2009): Outline and surface disruption in animal camouflage. Proc. R. Soc. Biol. Sci. Ser. B 276: 781-781. Vanni, S., Cimmaruta, R. (2010): Natrix cetti Gené, 1839. In: Fauna d’Italia. Reptilia, pp. 538-545. Corti, C., Capula, M., Luiselli, L., Razzetti, E. Sindaco, R., Eds, Edizioni Calderini de Il Sole 24 Ore Editoria Special- izzata S.r.l., Bologna. Vaughan, R. (1999): Provisional results from study of facial features as a means of individual identification in Natrix natrix. Br. Herpetol. Soc. Bull. 68: 39-46. Zuffi, M.A.L. (2008): Colour pattern variation in popula- tions of the European Whip snake, Hierophis viridifla- vus: does geography explain everything? Amphibia- Reptilia 29: 229-233.
Acta Herpetologica 14(2): 147-151, 2019 DOI: 10.13128/a_h-7754
Estimating abundance of the Stripeless tree-frog Hyla meridionalis by means of replicated call counts
Federico Crovetto, Sebastiano Salvidio, Andrea Costa* DISTAV, University of Genova, Corso Europa 26, I-16132 Genova, Italy. *Corresponding author. Email: [email protected]
Submitted on: 2019, 16th May; revised on: 2019, 22nd July; accepted on: 2019, 16th September Editor: Marco Mangiacotti
Abstract. The Stripeless tree-frogHyla meridionalis reaches its eastern-most European distributional limit in NW Ita- ly, and specifically in the Cinque Terre National Park. Here for two consecutive years, we estimated tree-frog popula- tion abundance by call surveys at 24 sites. Data were analysed in the framework of N-mixture open population mod- els based on repeated counts of calling males. The results obtained by this statistical approach were effective in esti- mating population size together with annual recruitment and survival. The tree-frog male population size remained constant between years and site abundance was inversely related with altitude. On the bases of these findings, our application of N-mixture models to tree-frog calling males was successful and is a promising cost-effective method to obtain long-term monitoring data on this species over large geographic areas.
Keywords. Abundance estimation, call surveys, Cinque Terre National Park, detection probability, N-mixture models.
The Stripeless tree-frog Hyla meridionalis Boettger, arboreal and highly secretive during daytime. The use of 1874 is found in North-western Africa (Algeria, Morocco PVC pipes may increase the probability of detection of and Tunisa), South-western Europe (Portugal, Spain, South tree-frogs (do Vale et al., 2018), however, in the CTNP France and North-Western Italy) and the Canary and the majority of the species’ reproductive habitats are on Balearic Islands (Sillero, 2010). On a portion of its distribu- private lands, and thus are not freely accessible (Romano tion range the species is considered introduced, i.e., Canary et al., 2014). Therefore, the monitoring technique select- and Balearic Islands (Sillero et al., 2014), and it is also pos- ed to estimate population size was based on nocturnal sibly introduced for other European regions (Recuero et auditory surveys of calling males, because of the species al., 2007). In Italy, the Stripeless tree-frog is common along highly distinctive mating call (Schneider, 1974; Márquez the Mediterranean coast of Liguria (NW Italy), from the et al., 2005). Call survey is a relatively efficient technique Province of Imperia to the province of La Spezia (Salvidio, for evaluating the distribution and diversity of anurans 2007). Apart from morphometric and distributional data (Dorcas et al., 2009). Therefore, calling surveys are fre- (Salvidio, 2007), little is known about the abundance and quently used in large-scale amphibian monitoring pro- dynamics of Stripeless tree-frog populations in Italy, and grammes (e.g., Anthony, 2002; Weir and Mossman, 2005; quantitative data on populations size should be obtained to Weir et al., 2005, 2009). However, the use of call surveys assess the species status and its ecological requirements, in for estimating population abundances and trends suffers particular near the species distribution limits, where a high of the same problematic issues recognized in the case of population fragmentation is expected (Gaston, 2003). repeated counts of individuals, because the detectability Although photo-identification of Stripeless tree-frogs of anuran calling males is < 1 (i.e., not all males are call- is possible (Crovetto unpublished data), the animals are ing in the same night; Schmidt and Pellet, 2005. Moreo-
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 148 Federico Crovetto, Sebastiano Salvidio, Andrea Costa ver, anuran mating call activities display high variation (from West to East): 6 streams and 18 artificial water res- in response to biotic and abiotic factors, that usually ervoirs in agricultural lands or urban settings (Table 1; remain unknown and difficult to model (Royle and Link, Fig. S1). 2005; Droege and Eagle, 2009). In fact, using raw counts All surveys began after sundown and after hear- of calling males or even scores derived from abundance ing the first tree-frog calls. In 2017 three nocturnal sur- indexes (i.e., indexes that group calling males by classes veys were performed, from the end of March to May, by of relative abundance; Weir and Mossman, 2005) without two operators that counted the number of males calling accounting for detection probability may lead to relevant at each site during a two minute period. In 2018, three bias in abundance and trend estimates (Schmidt, 2004; nocturnal surveys were performed, from the beginning Mazerolle et al., 2007). Therefore, to reliably estimate of May to the beginning of June, with the same observers population abundance, the information derived from and procedure of 2017. In addition in 2018, a fourth sur- raw counts of calling males should always be corrected vey was performed by a single operator that tallied call- for species-specific detection probabilities (Schmidt and ing males for 4 minutes. The asynchrony and the different Pellet, 2005; Royle and Link, 2005). Recently, specific tonalities of calls permitted to count with confidence the modelling approaches have been proposed for estimating minimum number of males per site that, in all cases, was anuran population abundances from the count of anuran ≤ 6 (Table S1). All sites were surveyed during the same calling indexes taking into account detection probabilities overcast or rainy night, but never during heavy showers (Royle, 2004a; Royle and Link, 2005). that could hinder a clear hearing of frog calls. Four cli- This study aimed to estimate the abundance of matic variables were obtained from the meteorological Stripeless tree-frog males together with some demo- station of Levanto: rainfall during the 24 h preceding the graphic parameters and ecological requirements in Italy, survey (RAIN), air temperature (TEMP), relative humid- at the eastern limit of the species distribution. Moreover, ity (RH) and wind speed (WIND), recorded during the we tried to establish a cost-effective monitoring protocol last hour of survey. These weather variables were select- to provide future population trends. Because of the rela- ed, because they are known to influence anuran calling tively small number of tree-frog males recorded per site, behaviour (e.g., Walls et al., 2011). Finally, for each site we had the opportunity to apply the open population three variables were considered: altitude above the sea generalization of Royle’s (2004b) N-mixture model (Dail level (ELEV), a categorical variable for the municipality and Madsen, 2011) to count data derived from call sur- of the site (CITY) and if the water body was a stream or veys. an artificial site (SITE). The eastern-most limit of the species’ range in Repeated count data were analysed using the Dail- Europe is the village of Riomaggiore (Province of La Madsen (2011) model, which is a generalization of the Spezia), in the Cinque Terre National Park (CTNP), a Royle’s (2004b) N-mixture model, capable of relaxing the protected area where the Stripeless tree-frog reproduces closure assumption by considering the population closed in streams and in artificial water tanks used for irriga- to immigration/births and emigration/deaths during a tion (Salvidio, 2007; Romano et al., 2014). In this area short period (i.e., three/four survey nights performed water streams display short and steep courses, with rela- each year), while considering the population demograph- tively long summer drying periods, due to the lack of ically open between years, in a robust design-similar precipitations (Olivari et al., 2013). Among many pos- approach. This model estimates four parameters, two of sible land use of rural areas, agriculture is the only one which are in common with the Royle’s (2004b) N-mix- in the CTNP, and from the sea level up to the hill tops ture original formulation: individual detection prob- vineyards and orchards are cultivated on strips of arable land, or “terraces”, sustained by dry-stone walls. Irriga- tion is provided by means of water stored in tanks, often colonised by amphibians (Olivari et al., 2013; Romano et Table 1. Continuous variables included in the N-mixture open al., 2014). The survey sites were selected during both day- population models (Dail and Madsen, 2011) used to estimate Hyla meridionalis abundance, in the Cinque Terre National Park. time and nocturnal preliminary surveys. During the day, streams and water reservoirs were inspected and selected Sample size Variable Description Mean (SD) min max as potential reproductive sites if adults, larvae or eggs of (N) some amphibian species were observed. During the night, ELEV Site altitude (m) - 24 35.04 (29.66) 8 83 sites were located by perceiving the calls of Stripeless TEMP Air temperature (°C) 7 17.14 (3.44) 12 21 three-frog males. In total 24 sites were surveyed in the WIND Wind speed (m/s) 7 3.57 (1.27) 2 5 municipalities of Levanto, Monterosso and Riomaggiore RH Relative humidity (%) 7 66 (13.55) 45 85 N-mixture models for Hyla meridionalis call counts 149
Table 2. Candidate N-mixture open population models (Dail and Madsen, 2011) used to estimate Hyla meridionalis abundance, ranked by AICc. γ = recruitment rate; λ= initial site abundance; p = individual detection probability; ω = survival; AICcWT = model weights. In model list t stands for time dependence. For covariate abbreviations see table 1.
Model Parameters AICc ∆AICc AICcWT λ(ELEV) p(.) ω(.) γ(.) 5 451.35 0.00 0.40 λ(ELEV) p(.) ω(ELEV) γ(.) 6 453.45 2.1 0.14 λ(CITY) p(.) ω(.) γ(.) 6 453.98 2.63 0.11 λ(ELEV) p(.) ω(CITY) γ(.) 6 454.95 3.61 0.07 λ(.) p(.) ω(.) γ(.) 4 454.96 3.61 0.07 λ(SITE) p(.) ω(.) γ(.) 5 455.05 3.70 0.06 λ(ELEV) p(.) ω(SITE) γ(.) 7 456.15 4.80 0.04 λ(.) p(TEMP) ω(.) γ(.) 5 456.81 5.46 0.03 λ(.) p(.) ω(SITE) γ(.) 5 456.93 5.59 0.02 Fig. 1. Effect of elevation on site specific abundance of Hyla meridi- λ(.) p(RH) ω(.) γ(.) 5 457.14 5.80 0.02 onalis in the Cinque Terre National Park, with 95% confidence λ(.) p(WIND) ω(.) γ(.) 5 475.54 6.20 0.02 intervals, obtained by N-mixture open population modelling (Dail λ(.) p(RAIN) ω(.) γ(.) 5 458.03 6.69 0.01 and Madsen, 2011). λ(.) p(.) ω(ELEV) γ(.) 5 459.15 7.80 0.01 λ(.) p(.) ω(CITY) γ(.) 6 459.35 8.01 0.01 λ(.) p(t) ω(.) γ(.) 11 483.07 31.73 0.00 be affected by climatic variables. Then we modelled the initial abundance as a function of site specific covariates (ELEV, CITY and SITE) or constant over sites. Finally, we considered the survival to be influenced by the same site ability (p) and mean initial abundance for each site (λ). covariates as abundance, or constant over sites. For each The Dail-Madsen (2011) model estimates two additional model we considered recruitment as constant. We ranked parameters: the recruitment rate (γ), comprehensive of all models with Akaike’s Informative Criterion corrected births and immigrations, and the apparent survival prob- for small samples (AICc). We conducted model selection ability (ω), comprehensive of deaths and emigrations. In and considered only models with ∆AICc > 2 (Burnham our study, we built models with Poisson error distribu- and Anderson, 2002). Modelling was conducted in the R tion, since Negative Binomial distribution could lead to environment with package Unmarked (Fiske and Chan- identifiability issues and may produce infinite abundance dler, 2011) and AICcmodavg (Mazerolle, 2017). estimates (Barker et al., 2017; Link et al., 2018). Further- In 2017 we counted a total of 131 male frog calls more, in order to avoid truncated estimates of abundance during three surveys (44; 45; 42; respectively), while in (Knape et al., 2018), we set the upper limit for integration 2018 we counted 129 male frogs during four surveys (33; (K) to 50 (i.e., we checked estimate stability at incremen- 37; 32; 27; respectively, Table S1). The global model had a tal values of K). We then began the model building pro- good fit (goodness-of-fit, P = 0.34; c-hat overdispersion = cedure by fitting a global model (i.e., the most complex 1.12, and visual inspection of residuals). Model building model on which other models are nested) and assessing procedure produced a total of 15 models (Table 2). The the fit of this model in two ways: i) by means of a Pear- most supported model included elevation as a covariate son chi-square test (MacKenzie and Bailey, 2004), using on the initial abundance, highlighting a negative effect of a parametric bootstrap procedure (5000 re-samplings), elevation (βELEV = -0.331; 95% CI = -0.59 to -0.08; Figure ii) by inspecting residuals (Knape et al., 2018). In order 1). The estimated mean frog abundance per site was 3.4 to avoid overfitting and creating too many models, deriv- (95% CI = 2.5 – 4.6). Individual detection probability for ing from the combinations of covariates for each of the this model was constant, and estimated as p = 0.53 (95% four parameters of the Dail-Madsen (2011) model, which CI = 0.42 – 0.63). Survival probability between years was can lead to uninformative and biologically unsound considered constant among sites and resulted ω = 0.71 models, we preferred to build fewer models in a step- (95% CI = 0.50 – 0.86). Finally, the recruitment rate was wise approach, considering one parameter at a time, and constant across sites γ = 0.11 (95% CI = 0.00 – 12.10). building biologically informative models. From this best model we also obtained, as a derived We proceeded modelling the detection probabil- parameter from the posterior distribution of the latent ity, considering it to be constant, time-dependent, or to abundance, the total abundance of surveyed sites, which 150 Federico Crovetto, Sebastiano Salvidio, Andrea Costa resulted of 89 frogs in 2017 (95% CI = 81 – 147) and 64 SUPPLEMENTARY MATERIAL frogs in 2018 (95% CI = 60 – 109). Our study showed that N-mixture modelling applied Supplementary material associated with this article to individual frog calls can be successfully used to esti- can be found at
Prelude and static models. Elsevier/Academic Press, Royle, A.J. (2004b): N-mixture models for estimat- Amsterdam, The Netherlands. ing population size from spatially replicated counts. Knape, J., Arlt, D., Barraquand, F., Berg, Å., Chevalier, Biometrics 60: 108-115. M., Pärt, T., Ruete, A., Żmihorski, M. (2018): Sensitiv- Royle, J.A., Link; W.A (2005): A general class of multino- ity of binomial N-mixture models to overdispersion: mial mixture models for anuran calling survey data. the importance of assessing model fit. Met. Ecol. Evol. Ecology 86: 2505-2512. 9: 2102-2114. Salvidio, S. (2007): Hyla meridionalis Boettger, 1874. In: Link, W.A., Schofield, M.R., Barker, R.J., Sauer, J.R. Fauna d’Italia – Amphibia. XLII, pp. 338-346. Lanza, (2018): On the robustness of N-mixture models. Ecol- B., Andreone, F., Bologna, M.A., Corti, C., Razzetti, ogy 99: 1547-1551. E., Eds, Calderini, Bologna. MacKenzie, D.I., Bailey, L.L. (2004): Assessing the fit of Schmidt, B.R., (2004): Declining amphibian populations: site-occupancy models. J. Agric. Biol. Environ. Stat. 9: the pitfalls of count data in the study of diversity, dis- 300-318. tributions, dynamics, and demography. Herpetol. J. Márquez, R., Moreira, C., do Amaral, J.P.S., Pargana, J.M., 14: 167-174. Crespo, E.G. (2005): Sound pressure level of adver- Schmidt, B.R., Pellet, J. (2005): Relative importance of tisement calls of Hyla meridionalis and Hyla arborea. population processes and habitat characteristics in Amphibia-Reptilia 26: 391-395. determining site occupancy of two anurans. J. Wildlife Mazerolle, M.J., Bailey, L.L., Kendall, W.L., Royle, J.A., Manage. 69: 884-893. Converse, S.J., Nichols, J.D. (2007): Making great leaps Schneider, H. (1974): Structure of the mating calls and forward: accounting for detectability in herpetological relationships of the European tree frogs (Hylidae, field studies. J. Herpetol. 41:672-689. Anura). Oecologia 14: 99-110. Mazerolle, M.J. (2017): AICcmodavg: model selec- Sillero, N. (2010): Modelling suitable areas for Hyla tion and multimodel inference based on (Q)AIC(c). meridionalis under current and future hypothetical R package version 2.1-1. https://cran.r-project.org/ expansion scenarios. Amphibia-Reptilia 31: 37-50. package=AICcmodavg. [Accessed on 22 January 2019] Sillero, N., Campos, J., Bonardi, A., Corti, C., Creem- Olivari, S., Romano, A., Salvidio, S. (2013): Anfibi e habi- ers, R., Crochet, P-A., Crnobrnja Isailovic, I., Denoël, tat acquatici nel Parco Nazionale delle Cinque Terre - M., Ficetola, G.F., Gonçalves, J., Kuzmin, S., Lym- Censimento e indirizzi per la conservazione. Edizioni berakis, P., de Pous, P., Rodríguez, A., Sindaco, R., Belvedere, Latina. Speybroeck,J., Toxopeus, B., Vieites, D.R., Vences, M. Priol, P., Mazerolle, M., Imbeau, L., Drapeau, P., Trudeau, (2014): Updated distribution and biogeography of C., Ramière, J. (2014): Using dynamic N-mixture amphibians and reptiles of Europe. Amphibia-Reptilia models to test cavity limitation on northern flying 35: 1-31. squirrel demographic parameters using experimental Walls, S., Waddle, J.H., Dorazio, R.M. (2011): Estimating nest box supplementation. Ecol. Evol. 4: 2165-2177. occupancy dynamics in an anuran assemblage from Recuero, E., Iraola, A., Rubio, X., Machordom, A., Louisiana, USA. J. Wildlife Manage. 75: 751-761. García-París, M. (2007): Mitochondrial differentiation Weir, L., Fiske, I.J., Royle, J.A. (2009): Trends in anuran and biogeography of Hyla meridionalis (Anura: Hyli- occupancy from northeastern states of the North dae): An unusual phylogeographical pattern. J. Bioge- American Amphibian Monitoring Program. Herpetol. ogr. 34: 1207-1219. Cons. Biol. 4: 389-402. Romano, A., Costa, A. Basile, M. Raimondi, R., Posillico, Weir, L.A., Mossman M.J. (2005): North American M., Scinti Roger, D., Crisci, A., Piraccini, R., Raia, P., Amphibian Monitoring Program. In: Declining Matteucci, G., De Cinti, B. (2017): Conservation of amphibians: A United States response to the global salamanders in managed forests: Methods and costs phenomenon, pp. 307-313. Lannoo, M.J., Ed, Univer- of monitoring abundance and habitat selection. For. sity of California Press, Berkeley, California, USA. Ecol. Manage. 400: 12-18. Weir, L.A., Royle, J.A., Nanjappa, P., Jung, R.E. (2005): Romano, A., Salvidio, S., Mongillo, D., Olivari, S. (2014): Modeling anuran detection and site occupancy on Importance of a traditional irrigation system in North American Amphibian Monitoring Program amphibian conservation in the Cinque Terre National (NAAMP) routes in Maryland. J. Herpetol. 39: 627- Park (NW Italy). J. Nat. Cons. 22: 445-452. 639. Royle, A.J. (2004a): Modeling abundance index data from anuran calling surveys. Cons. Biol. 18: 1378-1385.
Acta Herpetologica 14(2): 153-158, 2019 DOI: 10.13128/a_h-7755
AT-rich microsatellite loci development for Fejervarya multistriata by Illumina HiSeq sequencing
Yan-Mei Wang, Jing-Yi Chen, Guo-Hua Ding*, Zhi-Hua Lin ADI, College of Ecology, Lishui University, Lishui 323000, Zhejiang, China. * Correspondence author. Email: [email protected]
Submitted on 2018, 22th May; Revised on: 2019, 3rd June; Accepted on: 2019, 23rd August Editor: Emilio Sperone
Abstract. In our study, a total of 2561 sequences that contained microsatellite loci were found potentially to be used for primer design. Furthermore, Illumina HiSeq sequencing technology identified trinucleotide repeats and AT-rich repeats with the the highest proportion in our genomic DNA sequence library of Fejervarya multistriata. Eighteen new microsatellite loci of F. multistriata were isolated and we characterize these loci genotyping 48 individuals sam- pled from 3 populations in Lishui City, Zhejiang Province, China. Seventeen loci were polymorphic, with the number of alleles ranging from 2 to 11 within each population. The polymorphic information content, observed and expected heterozygosity ranged 0-0.845, 0-1.0 and 0-0.871, respectively. None of the loci was observed in linkage disequilib- rium. One locus (FMA294) was deviated from Hardy-Winberg equilibrium in each population separately and com- bined. These informative microsatellite loci will be applicable for conservation genetic studies of F. multistriata across varying scales from inter-individual to inter-population.
Keywords. Fejervarya multistriata, genome, microsatellite, next-generation sequencing, polymorphism.
INTRODUCTION 2012; Lü et al., 2013; Zhang et al., 2013; Sultana et al., 2014; Igawa et al., 2015; Song et al., 2017). Microsatellite DNA loci, also known as simple Fejervarya multistriata (Anura: Dicroglossidae) is sequence repeats, occur at thousands of locations within a species of frog, which is widely found in south of the the eukaryotic genome, and are highly variable and suffi- Yellow River in China and some countries (regions) in cient in nuclear genome (Ellegren, 2004; Wei et al., 2015; Southeast Asia, such as northern India, Vietnam and Shao et al., 2017). Therefore, microsatellite DNA loci are Myanmar (AmphibiaChina, 2018). The conservation widely applied as molecular markers in population genet- status of this species is listed as data deficient in IUCN ics, species identifying, genetic breeding and genetic map (AmphibiaChina, 2018). This species prefers to inhab- (Selkoe and Toonen, 2006; Abe et al., 2012; Wambulwa et it paddy field and still water and its ovulation remains al., 2016; Soulard et al., 2017). Thanks to the development active from April to September every year (Amphibi- of next-generation sequencing technology, both through- aChina, 2018). As a dominant amphibian species, F. mul- put and efficiency of developing microsatellite DNA has tistriata plays an important role in farmland ecosystem, increased with a decreased cost of sequencing process. and its population density in field has decreased due to In recent years, microsatellite DNA markers have been urbanization (Li et al., 2016). Meanwhile, environmen- quickly developed in many species at a low cost when tal degradation also threatens the survival of this species using the next-generation sequencing technology on Illu- (Othman et al., 2009). In previous studies, mitochon- mina HiSeq and Ion Torrent PGM platforms (Yang et al., drial D-loop sequences were used as molecular mark-
ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah 154 Yan-Mei Wang et alii ers to study phylogeography of F. multistriata popula- Biotech Ltd. Co., Shanghai, China). The PCR products were tions (Zhong et al., 2008). Twenty-one microsatellite loci, genotyped on an ABI 3730 sequencer (Applied Biosystems) and mainly including dinucleotide repeats, had been isolated following data were analyzed with GeneMarker v1.8 software. for the species, and only approximately 24% loci had AT Population genetic parameters for polymorphic loci such as number of alleles (N ), observed heterozygosity (H ), expected repeats (Chen et al., 2018). Here, we sequentially devel- A O heterozygosity (HE), polymorphic information content (PIC), P oped 18 new microsatellite DNA loci containing AT-rich values in Hardy-Weinberg equilibrium (HWE) tests and linkage repeats for F. multistriata by Illumina HiSeq sequencing. disequilibrium were calculated by Genepop 4.0 (Rousset, 2008), These new AT-rich microsatellite loci definitely would Cervus 2.0 (Marshall et al., 1998) and Fstat 2.9.3.2 (Goudet, be useful in examining genetic diversity and protecting 1995), respectively. genetic resources of F. multistriata.
RESULTS AND DISCUSSION MATERIAL AND METHODS Sequence data from Illunima HiSeq Forty-eight F. multistriata individuals (n = 16 for each pop- ulation) used in this study were collected by hand and net from We obtained a total of 6970707900 bp and 23235693 3 localities in Lishui City, Zhejiang Province, China, which were reads in a single sequencing run on Illunima HiSe- Lanshantou (LST, 119.7607°E, 28.36366°N), Baimashan (BMS, qTM using RAD-Tag. The distribution frequency of read 119.1337°E, 28.63823°N) and Jiulongshan (JLS, 118.8452°E, length for this species had a single peaks at approximately 28.39538°N), respectively. Our experimental procedures are 125 bp. In a total of 307793 reads with more than 125 bp, compliant with current laws on animal welfare and research in China, which are also specifically approved by the Animal 2561 reads contained microsatellite loci (0.83%). Research Ethics Committee of Lishui University (Permit No. Compared to a traditional library-based approach AREC-LU 2017-04). such as magnetic beads enrichment (Guo et al. 2015; Genomic DNA was extracted from toe muscle tissue of Chang et al. 2016), next-generation sequencing technol- one male F. multistriata from LST population using the DNeasy ogy is a more powerful approach to develop microsatel- Tissue Kit (Qiagen). The concentration of DNA sample was lite markers due to its efficiency and low cost. Sufficient measured by using a spectrophotometer at 260 and 280 nm and microsatellite sequences can be constructed in a genomic DNA sample was quantified on an agarose gel. A 200-400 bp DNA sequence library on Illunima HiSeqTM using RAD- sequencing library was constructed according to the manufac- Tag. This result agrees with recent studies on other anu- turer instructions (Illumina). This library was sequenced using an Illumina HiSeq 2500 Platform with RAD-Tag at Novogene ran species (Wei et al., 2015; Shao et al., 2017). Further- Bioinformatics Technology Co., Ltd (Beijing, China, http:// more, microsatellite obtained rate maybe higher on Ill- www.novogene.com/). The microsatellite primer pairs of F. unima HiSeq platform (e.g., 0.83% in our study) than on multistriata were designed using Primer Premier 3.0 software, Ion Torrent PGM platform (e.g., 0.32–0.57% in Igawa et which was used to check against potential primer dimers, hair- al., 2015). pin structures and the occurrence of mismatches. Parameters for designing the primers were set as follows: primer length ranged from 18 bp to 24 bp with 22 as the optimum; PCR prod- Frequency and distribution of microsatellite loci in the uct size ranged from100-280 bp; melting temperature ranged genome from 50 °C to 65 °C with 55 °C as the optimum annealing tem- perature; GC content ranged from 30% to 70% with 50% as The length of the microsatellite loci ranged from the optimum. Finally, thirty newly designed primer pairs were 12 to 33 bp (15.7 ± 5.2, mean ± SD). The microsatellite selected to synthesize and initially screened for microsatellite DNA loci included 5 motif types: dinucleotide repeats loci using total genomic DNA isolated from 6 F. multistriata individuals collected from the LST population. (36.20%), trinucleotide repeats (52.60%), tetranucleotide PCR amplification reactions were performed using a ther- repeats (9.64%), pentanucleotide repeats (0.90%) and mal cycler (T100, Bio-Rad, USA). The total volume of each PCR hexanucleotide repeats (0.66%) (Fig. 1A). The frequency mixture was 20 μL, containing 1 μL genomic DNA (100 ng/μL), distribution of the 5 motif types was different signifi- 10 μL Premix Taq (TaKaRa, Japan), 1 μL of each primer (10 cantly (G-test, G = 3085.9, df = 4, P < 0.001). The motif μM) and 6 μL double distilled H2O. The conditions of the PCR repeat number of microsatellite loci ranged from 4 to 16, amplification were as follows: 95 °C for 5 min, then 35 cycles at while 97.66% of the microsatellite loci had 4-12 motif 95 °C for 30 s, Ta (the optimal annealing temperatures, see Table repeats, and motifs with more than 12 repeats were only 1) for 30 s, 72 °C for 30 s, and a further extension at 72 °C for with a frequency of <1.0% (Fig. 1B). There were 4 dinu- 10 min. Twenty-six primer pairs were further selected due to successful amplification in the 6 individuals, and the forward cleotide motif types, and the main types were AC/GT primer was labeled with FAM or HEX fluorescent dye (Sangon (50.70%), AT (35.17%) and AG/CT (13.92%) (Fig. 1C), Microsatellite loci for F. multistriata 155
Fig. 1. Characterization of microsatellite loci in Fejervarya multistriata genome. (A) distribution of different repeat motif types of microsat- ellite loci; (B) number of different repeat motifs; (C) frequency distribution of 5 different repeat types based on different motif types. Num- ber represent number of sequence. respectively. There were 10 trinucleotide motif types, and gested that the frequency of the repeat type changed ran- the main types were AAT/ATT (60.65%) and AGG/CCT domly for each species and was species-specific. The fre- (13.51%) (Fig. 1C), respectively. There were 18 tetranu- quencies decreased when the repeat unit length in each cleotide motif types, and the main types were AAAT/ repeat type motif of F. multistriata increased (Fig. 1B), ATTT (43.32%), ACAT/ATGT (11.34%), AATT (10.12%) indicating that a relatively short repeat unit of microsat- and AGAT/ATCT (10.12%) (Fig. 1C), respectively. There ellites might be a main component in the genome of F. were 13 pentanucleotide motif types, and the main types multistriata. were AAAAT/ATTTT (39.13%) and AAATT/AATTT Our finding suggested that the ratio of the repeat (13.04%) (Fig. 1C), respectively. There were 6 hexanu- motifs with an AT content of approximately 56% in our cleotide motif types, and the main types were AAAAAT/ F. multistriata was similar to other reported anuran ATTTTT (58.82%) and ACTCCG/CGGAGT (17.65%) species (e.g., X. tropicalis, Xu et al., 2008; O. narina, H. (Fig. 1C), respectively. tigerinus and B. japonica, Igawa et al., 2015), suggesting The frequency of different repeat types of microsatel- that the AT content could be an important repeat unit in lites in F. multistriata was different from Xenopus tropi- anurans. The dominant repeat motif in the trinucleotide calis (Xu et al., 2008), Odorrana narina, Hoplobatrachus type of F. multistriata, (AAT/ATT repeat) was similar to tigerinus, and Buergeria japonica (Igawa et al., 2015). The the other four reported anuran species (Xu et al., 2008; dominant repeat type was trinucleotide in F. multistri- Igawa et al., 2015). However, other types of repeat motif ata, but dinucleotide in the last 4 anuran species (Xu et were different among these species. For example, F. mul- al., 2008; Igawa et al., 2015). Such results may be related tistriata had a higher frequency in AC/GT and AAAT/ to the different next-generation sequencing platforms ATTT repeats, but O. narina, H. tigerinus, B. japonica used in constructing sequence library. Since the library and X. tropicalis in AT and AGAT/ATCT repeats (Xu et of microsatellite sequences was constructed by Illunima al., 2008; Igawa et al., 2015). These results implied that HiSeq platform for F. multistriata, however, by Ion Tor- the accumulation rates of repeat motifs were maintained rent PGM platform for O. narina, H. tigerinus and B. in modern anurans, but skewed in a common ancestor japonica (Igawa et al., 2015). In addition, the results sug- (Igawa et al., 2015). 156 Yan-Mei Wang et alii
Table 1. Characterization of 18 microsatellite DNA markers developed for F. multistriata. Size range: size range of fragment; bp: base pair; Ta,: annealing temperature of primer pairs; Na: number of alleles; Ho: observed heterozygosity; HE: expected heterozygosity; HWE: Hardy- Weinberg equilibrium; PIC: polymorphic information content; bold: significant deviation from HWE after Bonferroni correction (P < 0.05).
Locus T Size range Primer sequences (5’-3’) Repeat motif a N H H P PIC (GenBank #) (°C) (bp) A o E HWE FMA102 F: GCACTGTAGAGCACTGGATTC (TA)16 53 129-219 13 0.4792 0.7592 1.000 0.72 MG744293 R: GAGCGTCATAGGGGTCAAATAG FMA349 F: CACTCATGTTATCACTCTACTCTC (TAT)11 53 156-237 11 0.3333 0.8145 1.000 0.782 MG744294 R: CCTCCTACCTCTTGTACTAAATTG FMA117 F: ACTTGAGTCTATTCTATTCTGCTG (ATA)7 53 149-158 4 0.4167 0.5893 0.996 0.495 MG744295 R: ACTGCTGCTCTGATCTCTATG FMA402 F: AGACATTACCTTAAAGCCATAGTG (AGAT)6 53 189-201 4 0.4583 0.6090 0.975 0.538 MG744296 R: CTTCTGACATGACCTGTTCTTC FMA041 F: CCAGGAGGATTCTAGTGACAG (AT)12 53 152-192 12 0.3958 0.8169 1.000 0.789 MG744297 R: ATGAAGGCAAGAGCAATGTAC FMA466 F: GGTGCCACTGTCTTAACTATCC (TTTTTA)4 53 199-205 2 0.3125 0.4086 0.975 0.323 MG744298 R: AGTCCAATCAAGTCCAATTCAAAC FMA294 F: GTCCTCCTACCTCTTGTACTG (ATA)10 53 187-238 6 0.9375 0.6515 < 0.01 0.586 MG744299 R: CGAATGAGAACCTTCACAGAC FMA302 F: TCCGACCTCTTGAAACTGTATTG (ATA)10 53 205-259 13 0.6042 0.8680 1.000 0.845 MG744300 R: AGGATCACCACTAGGAGCATC FMA355 F: TATGACCACAGTCTAGCATCC (AAAT)5 58 158 1 – – na 0 MG744301 R: CTCCAGTAGTTATCACCTTCTTG FMA188 F: CCTCTTGTGTTGGTGTATTTCTG (AAT)8 63 209-215 3 0.6042 0.5340 0.170 0.434 MG744302 R: TTATGCTTGTGTTCTGGTCATTC FMA231 F: GCTGCTGCATGATAGTGTCTC (TAT)8 53 155-185 10 0.7917 0.8353 0.992 0.805 MG744303 R: TGATGTCTGATGGTCGTCCTG FMA072 F: TGCAGTAGACATCGGAGTTG (TA)13 53 209-237 10 0.6875 0.7934 0.998 0.757 MG744304 R: GCCTCTCTCATCTTATTAAGTGG FMA140 F: TTCATTGTGCCAAGTGTAACG (ATT)7 63 153-165 3 0.1875 0.2254 0.927 0.206 MG744305 R: TAACAAAGAGGTCATCACTAATCC FMA403 F: GCGTGGATCGTTATTGAAGTG (ATTT)6 63 193-197 2 0.2708 0.2945 0.858 0.249 MG744306 R: GGTGACCTAATGTGAAATTCCTG FMA116 F: CTCCCTAACTATTGTAAAGCACTG (ATA)7 55 165-195 8 0.5208 0.7825 1.000 0.743 MG744307 R: ATTATAGATGGAAGCAACAGGAAC FMA399 F: TTCAGGCTACAGGCATTACAG (AATA)6 55 168-228 4 0.2708 0.4476 1.000 0.416 MG744308 R: ATAAGGGTGTTCTGCTAAATCAAG FMA269 F: AATGCTTGCAGAACTATTCACAC (TAT)9 55 177-192 6 0.2500 0.6732 1.000 0.616 MG744309 R: TACGGCGGTCCTAAGATGG FMA139 F: GATTGATGGATTGATGATGGACTG (ATG)7 55 177-189 5 0.6042 0.6535 0.813 0.584 MG744310 R: AATGTTCAAGATGGACGAATTACC
Characterization of microsatellite loci for each locus were shown in Table 1. The aN , PIC and heterozygosities (HO and HE) ranged from 1 to 11 (4.815 Twenty-six primer pairs were used to successfully ± 2. 699, mean ± SD), 0 to 0.845 (0.549 ± 0.240, mean amplify genomic DNA of F. multistriata from LST popu- ± SD), 0 to 1.0 (0.452 ± 0.246, mean ± SD), 0 to 0.871 lation. Of the 26 pairs, 18 pairs generated target bands, (0.571 ± 0.237, mean ± SD) within each population, and the other 8 pairs generated non-target bands. Finally, respectively. No significant linkage disequilibrium was a total of 18 primer pairs were characterized, of which 17 observed after Bonferroni correction for multiple tests were polymorphic. The genomic sequences containing (all P > 0.05). Of the 18 loci, one (FMA294) deviated sig- a microsatellite locus, which were used to design these nificantly from HWE testing each population separately primers, were deposited in GenBank (accession num- and combined (all P < 0.05; Table 1), which indicated ber: MG744293–MG744310). The information of primer that null alleles may be present in this locus (Song et al., sequences, repeat motifs, Ta, NA, PIC and heterozygosity 2017). The locus was assessed to contain moderately high Microsatellite loci for F. multistriata 157 polymorphism degree when the value of PIC was larger ing: a case study of the endangered Odorrana narina than 0.5 (Song et al., 2017). Overall, 11 loci had high pol- complex of frogs. J. Hered. 106: 131-137. ymorphism degree (PIC > 0.5), 6 loci had low polymor- Li, B., Zhang, W., Shu, X.X., Pei, E.L., Yuan, X., Sun, phism degree (PIC < 0.5). These informative microsatel- Y.J., Wang, T.H., Wang, Z.H. (2016): The impacts of lite loci will be applicable for conservation genetic studies urbanization on the distribution and body condition of F. multistriata across varying scales from inter-individ- of the rice-paddy frog (Fejervarya multistriata) and ual to inter-population. gold-striped pond frog (Pelophylax plancyi) in Shang- hai, China. Asian Herpetol. Res. 7: 200-209. Lü, Z., Li, H., Liu, L., Cui, W., Hu, X., Wang, C. (2013): ACKNOWLEDGMENTS Rapid development of microsatellite markers from the large yellow croaker (Pseudosciaena crocea) using next We thank Yao-Fei Yu, Zhi-Qiang Chen for their help generation DNA sequencing technology. Biochem. in field work. This study was funded by National Science Syst. Ecol. 51: 314-319. Foundation of China (31500308), Zhejiang Provincial Marshall, T.C., Slate, J., Kruuk, L.E.B., Pemberton J.M. Natural Science Foundation of China (LQ16C040001), (1998): Statistical confidence for likelihood-based Key Research Projects of Lishui City (SH2017001), Zheji- paternity inference in natural populations. Mol. Ecol. ang Science and Technology Innovation Program for Col- 7: 639-655. lege Students (2018R434003, 2019R434006). Othman, M.S., Khonsue, W., Kitana, J., Thirakhupt, K., Robson, M. G., Kitana N. (2009): Cadmium accumu- lation in two populations of rice frogs (Fejervarya lim- REFERENCES nocharis) naturally exposed to different environmental cadmium levels. B. Environ. Contam. Tox. 83: 703-707. Abe, H., Hayano, A., Inoue-Murayama, M. (2012): Foren- Rousset, F. (2008): Genepop’007: a complete re-imple- sic species identification of large macaws using DNA mentation of the genepop software for Windows and barcodes and microsatellite profiles. Mol. Biol. Rep. Linux. Mol. Ecol. Resour. 8: 103-106. 39: 693-699. Selkoe, K.A., Toonen, R.J. (2006): Microsatellite for ecolo- AmphibiaChina (2019): The database of Chinese amphib- gists: a practical guide to using and evaluating micro- ians. Kunming Institute of Zoology (CAS), Kunming, satellite markers. Ecol. Lett. 9: 615-629. Yunnan, China. Available from: http://www.amphibi- Shao, W.W., Lin, Z.H., Fan, X.L., Ding, G.H., Wei L. achina.org/ [accessed on 1 March 2019] (2017): Characterization of 25 microsatellite loci for Chang, Y.C., Li, S.H., Lin, H.Y., Chen, S.L., Chang, M.H. the Guenther’s frog Hylarana guentheri. Conserv. (2016): Development of 22 polymorphic microsatellite Genet. Resour. 10: 31-33. markers for Taipei grass frogs (Hylarana taipehensis). Song, W., Zhu, D.M., Lv, Y.F., Wang, W.M. (2017): Isola- Amphibia-Reptilia 37: 117-120. tion and characterization of 37 polymorphic microsat- Chen, Y.D., Li R., Wei, L., Lin, Z.H., Li, Y.D., Ding, G.H. ellite loci of Megalobrama hoffmanni by next-genera- (2018): Development and characterization of 21 tion sequencing technology and cross-species amplifi- microsatellite loci for Fejervarya multistriata (Anura: cation in related species. J. Genet. 96: 39-45. Dicroglossidae) using genome-wide data. Microsatel- Soulard, L., Mournet, P., Guitton, B., Chaïr, H. (2017): lite records for volume 10, issue 1. Conserv. Genet. Construction of two genetic linkage maps of taro Resour. 10: 127-140. using single nucleotide polymorphism and microsat- Ellegren, H. (2004): Microsatellites: simple sequences ellite markers. Mol. Breeding. 37: 37. with complex evolution. Nat. Rev. Genet. 5: 435-445. Sultana, N., Igawa, T., Nozawa, M., Islam, M.M., Hasan, Goudet, J. (1995): FSTAT (Version 1.2): a computer pro- M., Alam, M.S., Khan, M.M., Sumida, M. (2014): gram to calculate F-statistics. J. Hered. 86: 485-486. Development and characterization of 27 new micros- Guo, K., Ding, G.H., Sun, Y.Y., Lu, H.L. (2015): Isola- atellite markers for the Indian bullfrog Hoplobatrachus tion and characterization of microsatellite loci in the tigerinus and its congeneric species. Genes Genet. northern grass lizard, Takydromus septentrionalis Syst. 89: 137-141. (Squamata: Lacertidae). Conserv. Genet. Resour. 7: Wambulwa, M.C., Meegahakumbura, M.K., Chalo, R., 583-584. Kamunya, S., Muchugi, A., Xu, J.C., Liu, J., Li, D.Z., Igawa, T., Nozawa, M., Nagaoka, M., Komaki, S., Oumi, Gao, L. M. (2016): Nuclear microsatellites reveal the S., Fujii, T., Sumida, M. (2015): Microsatellite marker genetic architecture and breeding history of tea germ- development by multiplex Ion Torrent PGM sequenc- plasm of East Africa. Tree Genet. Genome. 12: 11. 158 Yan-Mei Wang et alii
Wei, L., Li, L.M., Chen, J.P., Shao, W.W., Lin, Z.H. (2015): Characterization of 14 microsatellite loci for the pan- oriental narrow-mouthed toad, Microhyla ornata (Amphibia: Microhylidae). Conserv. Genet. Resour. 7: 33-35. Xu, Z.K., Gutierrez, L., Hitchens, M., Scherer, S., Sater, A. K.,Wells, D.E. (2008): Distribution of Polymorphic and Non-Polymorphic Microsatellite Repeats in Xeno- pus tropicalis. Bioinform. Biol. In. 2: 157-169. Yang, T., Bao, S.Y., Ford, R., Jia, T.J., Guan, J.P., He, Y.H., Sun, X.L., Jiang, J.Y., Hao, J.J., Zhang, X.Y., Zong, X.X. (2012): High-throughput novel microsatellite marker of faba bean via next generation sequencing. BMC Genomics 13: 602. Zhang, L.N., Peng, J., Li, X. J., Liu, Y.L., Cui, C. J., Wu, H., Wu, R.N., Tian, P.P., Li, Y. (2013): Development of 27 trinucleotide microsatellite markers for Saccharina japonica using next generation sequencing technol- ogy. Conserv. Genet. Resour. 6: 341-344. Zhong, J., Liu, Z.Q.,Wang, Y.Q. (2008): Phylogeography of the rice frog, Fejervarya multistriata (Anura: Rani- dae), from China based on mtDNA D-loop sequenc- es. Zool. Sci. 25: 811-882.
Finito di stampare da Logo s.r.l. – Borgoricco (PD) – Italia Cover: Top left: P. latastei from Ponza Island, male with spotted color pattern . Photo by Jean-Michel Delaugerre. Top right: P. latastei from Ponza Island, female with “quasi” concolor pattern. Photo by Jean-Michel Delaugerre. Bottom left: P. latastei from Ponza Island, female with “quasi” concolor pattern. Photo by Jean-Michel Delaugerre. Bottom right: P. latastei from Ponza Island, male from with reticulate color pattern. Photo by Jean- Michel Delaugerre.
© 2019 Firenze University Press Università degli Studi di Firenze Periodicità: semestrale Firenze University Press ISSN 1827-9643 (online) via Cittadella 7, 50144 Firenze, Italy ISSN 1827-9635 (print) http://www.fupress.com/ Registrata al n. 5450 del 3.11.2005 E-mail: [email protected] del Tribunale di Firenze December 2019 Vol. 14 – N. 2 ACTA HERPETOLOGICA 71 81 89 147 123 129 135 141 109 117 ,
uni s , , aime O
, J uez uca uez
, g L g , idha R limani omín odrí hiara , S R oberto D C - -
s R erreira cio s , be 14 – N. 2 2019 Vol. December F
a g ahar iller orti e C T ore
Z C S ,
o D . oberto
L R A
arcela
, e M g fenthouraki artín , aolo laudia rottini uetzaly ietro or S arra M C P C s
in
B , é , J , P , Q L s ier s .
o yro omero ini A p , P iva
ua ta R elica , J öhme
s V ia gg g a H - , S B o ía g - n ia g errera s
L C B g hi A H
uez
,
rte , Z g by Illumina HiSeq Fejervarya multistriata by Illumina HiSeq afili ular an , O alvatore g
g P eza M arta s s
, S in odrí ú M ndrea M olf s R . D e A hzaoui , illafañe
, A g V , J
a , W otto ua ván L anuela ar I ss
uiz H anayioti s lia
acchi u - alem M é R g s ha - er R , S ueno , P C alvidio u p ti
alvo g uo s B , V
S o M a ta uez
iou
s o R g , G o C g u g
, L ie
úñez M oberto arim - eor hen tiano iachello R D l N odrí s eña alvatore K - (Bedriaga, 1879) of the Western Pontine Islands (Italy) raised raised Pontine Islands (Italy) 1879) of the Western (Bedriaga, ouden , , E C R uez ,
G ario G
P i , , S M g eba ini Y M - la iccardo erre orti
, adi leni g , S e -G R EL g aino C E
C odrí , in
imone L D
,
R l - s , J an onzalo - araone E , S alenzuela elau g
F ss V hi D a it
, G
g vila an anna rovetto laudia iacalone arcía A orti H P. P. enczuk afaela aballero Á
avvide
C C S
l un C R
C G S G ean W
ecio co S , L E
, s s
ei -J anuel R ,
M M oni tina -
ch
s g s s ichel iovanni abriele abriele añete laudia ri o ndrea edouane o erónica an nrico ui ederico rance ablo anteli C G L V Estimating abundance of the Stripeless tree-frog Hyla meridionalis by means of replicated call counts F microsatellite loci development for AT-rich Y F ) with in the dorsal pattern of the Sardinian grass snake (Natrix natrix cetti Variability notes on its ecology E 153 sequencing Visible Implant Alphanumeric (VIA) as a marking method in the lesser snouted method snouted marking a as lesser the (VIA) in Alphanumeric Implant Visible treefrog Scinax nasicus A confirmed population of the Javelin sand boa, Morphological variation of the newly Italy in Sicily, Eryx jaculus (Linnaeus, 1758) (Serpentes, Erycidae) G tortoise Gopherus flavomarginatus in Mexico Hematological parameters of the Bolson C P on region, with comments in the Tensift Occurrence of Batrachochytrium dendrobatidis its spreading in Morocco R in skull morphology of two closely related Ontogenetic and interspecific variation (Anura: Bufonidae) species of toad, Bufo bufo and B. spinosus Coping with aliens: how a native gecko manages to persist on Mediterranean islands on Mediterranean islands how a native gecko manages to persist Coping with aliens: despite the Black rat? M field reptiles: insights from a long-term as a technique for marking fossorial PIT-Tags 101 wiegmanni study of the amphisbaenian Trogonophis Podarcis siculus latastei Podarcis siculus lizards and a brief taxonomic overview of Podarcis to the species rank, G of a Mediterranean lizard limited impact on the sprint performance Substrate type has a P CONTENTS C B C
15,00 € acta herpetologica acta